Patent Application
20150199743
Managing or otherwise processing the vast amount of information available in electronic trading of products can be difficult. In conventional web-based trading platforms, such difficulty to lead to poor engagement and churn of end-users.
The accompanying drawings are an integral part of the disclosure and are incorporated into the subject specification. The drawings illustrate example embodiments of the disclosure and, in conjunction with the present description and claims, serve to explain, at least in part, various principles, features, or aspects of the disclosure. Certain embodiments of the disclosure are described more fully below with reference to the accompanying drawings. However, various aspects of the disclosure can be implemented in many different forms and should not be construed as being limited to the implementations set forth herein. Like numbers refer to like elements throughout.
The present disclosure recognizes and addresses, in at least certain aspects, the issue of management of the vast amount of information available in web-based trading of products. Adequate management of such information can improve the usually poor user experience in web-based trading of products that can arise from exposure to the vast amount of information associated with such trading The disclosure provides, in certain embodiments, platforms, systems, devices, techniques, and computer-program products for content-based recommendations for trading of products. As an example, products in the present disclosure can include vehicles, computer products, firearms, articles of clothing, jewelry, consumer electronics, yard appliances, construction machines and equipment, aircraft, boats, office equipment, furniture, manufacturing equipment, packaging equipment, kitchen equipment, appliances, combinations of the foregoing, related products and components, or the like. It should be appreciated that while the disclosure is illustrated via embodiments, features, and the like, in which the products are or include vehicles, the features of the disclosure can be applied to other products as exemplified herein. The content-based recommendations in accordance with aspects of this disclosure can be utilized in organization-to-organization trading (such as trading amongst car dealerships) and can permit an organization (e.g., a dealer) to buy and/or sell products by matching the organization to suitable inventory and dealers. In certain implementations, embodiments of the disclosure can provide a flexible and scalable content-based hybrid recommendation platform that can generate a complementary set of dual recommendations for vehicles and dealership networks from a buying and selling perspective. More specifically, such a set is complementary in that the group of recommendations that is generated for vehicles (or other type of products) complements the group of recommendations that is generated for dealer networks (or other type of networks of product traders), and vice versa. Accordingly, a vehicle recommendation and a dealer network recommendation that are generated complement each other. In addition, the generated recommendations are dual with respect to buying and selling in that a group of recommendations that is generated includes two subgroups of recommendations: A subgroup of buying recommendations and a subgroup of selling recommendations. One or more of the generated recommendations can permit management of business operations of an organization, e.g., a dealer or the administrator of the content-based recommendation platform.
As described in greater detail below, a content-based recommendation platform and/or techniques in accordance with the present disclosure can uniquely leverage or otherwise utilize a business goal optimization module and several hybridization techniques to address multiple business operations (which also may be referred to business use cases or operational scenarios). For instance, offer acceptance rate on a portal or trading interface (e.g., a website configured by the administrator of such a platform). In certain embodiments, the content-based recommendation platform can include a profile learning module, a business goal optimization module, and a recommendation module. The profile learning module can compose or otherwise determine buying and/or selling profiles of an organization (e.g., a dealer) or an agent thereof. At least a portion of such profiles can be categorized as implicit or explicit. The business goal optimization module can determine (e.g., extract) a vehicle and/or dealership ranking model(s) that can be utilized or otherwise leveraged to optimize a business goal of the administrator of the content-based recommendation platform. Such model(s) may be referred to as rating model(s) and can include a model utilized or otherwise leveraged to produce a satisfactory (e.g., maximal or nearly-maximal) offer acceptance rate on a web-based trading platform for a product, such as a vehicle. It should be appreciated that in certain embodiments, the profile described herein can include various attributes representative or otherwise indicative of trading or otherwise commercial behavior of an organization. For instance, the profile (which may be referred to as a organization profile) can include or otherwise convey which vehicles a dealer turns, dealer's inventory (e.g., vehicles in the dealer's lot and/or vehicle of the dealer in stock), the inventory the dealer should stock, the vehicles that are moving into the area in which the dealer is located, the dealer's buying preferences (e.g., vehicles, counter parties, price range, condition, etc.), the dealer's selling patterns (vehicles, counter party types and identities, price range and flexibility, condition, etc.), recent activity, historical activity, peer and market activity, arbitrage opportunities and openness to arbitrage (e.g., geographic opportunities). Such rich behavior information for an organization can be leveraged herein to provide product recommendations for purchase of a product (e.g., a vehicle) for purchase for a trading organization, and to identify one or more organizations from which the recommended product can be purchased. In addition, in at least certain aspects, at least a portion of the trading behavior information can provide recommendations of a product for sale for the trading organization, wherein such a product for sale is included in a product inventory of the trading organization. Moreover, the recommendation of the product for sale can convey a network of one or more second trading organizations configured to purchase the second product.
The content-based recommendation platform and recommendation techniques of the disclosure can permit increasing the visibility of products (e.g., vehicles) that are simultaneously relevant to the user of a trading platform that leverages the content-based recommendation platform and, for example, maximize the offer acceptance rate in the event of an offer. As it can be appreciated, the relevance and ranking of a recommended product (e.g., a vehicle) can be business-use-case dependent. The disclosed content-based recommendation platform and associated recommendation techniques can support or otherwise permit implementation of multiple evaluation/rating procedures that contemplate specific business operation scenarios or objectives. In addition, such a content-based recommendation platform and techniques can provide sets of dual, complementary recommendations for products (e.g., vehicles) and organization (e.g., dealership) networks from a buying and selling perspective. It should be appreciated that the content-based recommendation platform and recommendation techniques in accordance with aspects of this disclosure can permit leveraging or otherwise utilizing a large volume of transaction information to the benefit of buyers and sellers, while providing a superior user experience. Therefore, in one aspect, user retention can be increased in a trading platform or an organization that operates or otherwise administers the content-based recommendation platform.
In certain embodiments of the content-based recommendation platform 110, the profile learning module can aggregate and/or generalize all or some of the information (e.g., data, metadata, and/or signaling) representative of organization preferences (e.g., user/dealership preferences) in order to construct organization profiles (e.g., user profiles, such as dealership profiles), which can include implicit buying and/or selling profiles, and/or explicit buying and/or selling organization profiles (e.g., user profiles, such as dealership profiles). It should be appreciated that the transaction information that is accessed or otherwise acquired can embody or otherwise include the content that permits generating recommendations of product for purchase or sale in accordance with aspects described herein.
In certain embodiments, the implicit buying and selling profiles can include (but are not limited to including) the following information: (A) Favorites “Year-Make-Model,” traded directly with another car dealership (e.g., traded outside of a vehicle trading platform for the trading of items between dealers); traded at wholesale auction and/or wholesale online timed sale (e.g., Online Vehicle Exchange (OVE), and/or other web-based or brick-and-mortar market places (such as auctions); searched or viewed on inventory trading/direct sales website (buying only, for example); made offers on inventory trading/direct sales website (buying only, for example); and/or placed bids on inventory trading/direct sales website (buying only, for example). (B) Favorites “price-range/odometer/car category (optional, for example)”, traded directly (e.g., outside of a vehicle trading platform) with another car dealership; traded at Wholesale Auction and/or Wholesale Online Timed Sale (such as OVE); searched or viewed on inventory trading/direct sales website (buying only, for example); made offers on inventory trading/direct sales website (applicable to buying profile, for example); and/or placed bids on inventory trading/direct sales website (applicable to buying profile, for example). (C) Fastest inventory turn Year-Make-Model. (D) Fastest inventory turn price-range/odometer/car category (optional, for example). (E) Slowest inventory turn Year-Make-Model. (F) Slowest inventory turn price-range/odometer/car category (optional, for example).
The explicit buying and selling profiles can include (but are not limited to including) the following information: (A) Saved searches: any filtering vectors saved by the user/dealership on inventory trading/direct sales website using available search criteria, such as year, make, model, trim, odometer, retail price range, distance, location, in personal network, inventory trading/direct sales website members, marked “ready to move”, marked in an event sales, body type, engine type, fuel type, transmission, equipment (standard, optional, aftermarket), dealer type, door, free form search text, . . . ). It should be appreciated that, in certain examples, saved searches may be applicable to buying profiles only. (B) User ratings: simple binary ratings and/or symbolic rating mapped to a numeric scale, made by the user (buyer) on inventory trading/direct sales website. (direct feedback, for example). (C) Vehicles marked as available for sale under specific trade condition(s) (e.g., “ready to move” vehicles). It should be appreciated that, in certain examples, information indicative or otherwise representative of marking vehicles in such a manner may be a feature applicable to selling profiles only. (D) Vehicles placed in inventory trading and/or direct sales website for a predetermined sale event (e.g., sales at specific time of the year, month, week, or day. It should be appreciated that, in certain examples, information indicative or otherwise representative of placement of vehicles in such a manner may be a feature applicable to selling profiles only.
As illustrated and described herein, the content-based recommendation platform 110 can include a recommendation module 120 that can support or otherwise utilize multiples profiles per organization (e.g., a user, such as a car dealership) corresponding to different operational scenarios (also referred to as business use cases). In certain implementations, the recommendation module 120 can utilize or otherwise leverage hybridization techniques to combine organization profiles (e.g., buying profiles, selling profiles, or a combination thereof) and generate one or multiple recommendation types (such as different recommendations for different scenarios, user stated preferences, and/or goals).
In certain embodiments, at least one profile (e.g., one profile, two profiles, more than two profiles, or each profile) can contain a set of search vectors weighted using the number of occurrence in the information (e.g., data and/or metadata) that is accessed by the content-based recommendation platform 110. The aggregation can be performed on any time period, such as the last 30, 45, 60, or 90 days of information (e.g., data and/or metadata), depending on the profiles considered. In some cases (e.g., Favorites “Year-Make-Model” traded directly with another dealership), forecasted data of these profiles can be used to capture seasonality effects and longer periods are then considered based upon data availability. The maximum number of vectors per profile can be limited (typically to a number from 10 to 20, for example) based on the processing hardware and/or other types of computing resources available.
In addition or in the alternative, organization profiles (e.g., user profiles, such as car dealership profiles) in accordance with the disclosure can be augmented, thus recommending additional profile vectors in order to overcome potential over-specialization effects and to promote or insure the diversity (e.g., novelty and serendipity) of the trading recommendations that are generated. In certain embodiments, the organization profiles can be augmented by using collaborative filtering techniques (e.g., finding the “k-nearest neighbors” organizations (e.g., users, such as car dealerships) and/or by generating new profiles via user-user and/or item-item filtering. In one example, the additional profile vectors generated using collaborative filtering can be specifically identified, so they can be processed accordingly by the recommendation module 120 during a hybridization phase of the recommendation processes in accordance with this disclosure.
As illustrated and described herein, the content-based recommendation platform 110 includes a business goal optimization module 150 that can extract a product (e.g., a vehicle and organization ranking model(s) (e.g., dealership ranking model(s)) that can be leveraged or otherwise utilized for optimizing or otherwise rendering satisfactory one or more business goals of the organization that administers (e.g., develops, deploys, leases, owns, or the like) the content-based recommendation platform 110. Multiple models are assigned to each User (e.g., car dealership) individually, aiming at optimizing or otherwise satisfactorily achieving multiple business goals and enabling multiple and personalized recommendations for a specific profile vector.
In certain embodiments, the primary business goal of the content-based recommendation platform 110, or an organization that operates or otherwise administers such a platform, can be to maximize the offer acceptance rate on trading platform (e.g., a website) of such the organization. As an illustration, the inventory trading/direct sales website offers data (offer accepted/rejected/pending, offer amount, . . . ) can be first joined with some or all of the data available about the products (e.g., vehicles), the seller organizations (e.g., seller dealerships), and/or the buyer organizations (e.g., the buyer dealerships). Such joined data can be processed using machine-learning techniques to extract the relevant factors driving the acceptance rate. The machine-learning techniques can include, for example, multi-layer neural networks and logistic multi-variable/multinomial regression analysis, naïve Bayes classifier, support vector machine (SVM), perceptron, linear discriminant analysis, quadratic classifier, and so forth. More specifically, yet not exclusively, logistic regression analysis can be implemented via the machine-learning techniques. It should be appreciated that, in certain embodiments, the specific machine-learning techniques relied upon in the present disclosure can be retained in the one or more rating model(s) 154. Without intending to be bound by theory, simulation, or modeling, the machine-learning techniques can be applied as described in DeMaris, “A Tutorial in Logistic Regression,” Journal of Marriage and the Family 57 (1995): 956-968. An example of a joined data table used for logistic multi-variable regression analysis in accordance with this disclosure is shown in
An example of a generic expression for the offer acceptance rating (F, which is a real number) can be formulated as:
Where p/(1+p) represents a probability of acceptance, p is a real number, and p=exp (ε+Σnβn*αn), ƒ is a real number and ƒ=Σmδm*θm. It should be appreciated that other formal expressions for the offer acceptance rating F can be defined or otherwise contemplated. The expression in Eq. (1) and/or other expressions for an offer acceptance rating can be included in one or more rating models 154. In Eq. (1), the coefficient ε is the intercept point of the multivariable regression model and some optional offsets (e.g., offsets that permit accounting for pricing assumptions); {αn} can embody or can include a list of N relevant parameters (e.g., attributes of the product (such as a vehicle), seller, buyer and the combination of buyer and seller); {β} is a list of N weight coefficients determined by the multi-variable regression analysis. Here, n is an index adopting natural number values and N is a natural number. In addition, ƒ is a custom function, which can be referred to as a “boost” function and, as described herein, is defined using a list of M parameters {δm} and its respective weight coefficient list {θm} where m is an index adopting natural number values and M is a natural number. It should be appreciated that, in certain embodiments, the parameters {δm} and the weight coefficients {θm} can be utilized or otherwise leverage to model additional or other business rules associated with trading, and can be selected or otherwise determined without the reliance on the machine-learning techniques. Machine-learning techniques as described herein can be are utilized to identify or otherwise determine a group of relevant parameters {αn}, which may be referred to as the “predictors,” and generate an estimated value of the parameters {βn} which are the respective weights for each of the group of relevant parameters {αn}. More specifically, yet not exclusively, the predictors and their respective beta weights of the logistic regression analysis can be determined concurrently using a combination of “maximum likelihood” estimation and an iterative process (for example using Newton's method), where a pseudo-R2 technique can be utilized to assess the quality of fit of the resulting model. The parameters {αn} and {βn} so determined permit determining a value of p as defined above. The value of p in conjunction with a computation of the boost function ƒ then permit determining a value of Gamma according to Eq. (1). In certain embodiments, N=12 and M=2, and examples of the relevant parameters {αn} (e.g., attributes) and {δm} can include one or more of the following, individually or in any combination:
In addition, regarding the boost function ƒ, the list {δn} (for N=2) can include two elements:
It should be appreciated that in certain implementations, other relevant parameters {αn} (e.g., attributes) can be leveraged or otherwise utilized to represent formally the offer acceptance rate.
In certain embodiments, the content-based recommendation platform 110, or an organization that operates or otherwise administers such a platform, can evaluate the commercial performance of a specific product category versus a market. In addition or in other embodiments, the content-based recommendation platform 110 can directly provide market insight to its customers. As an illustration, the performance versus the market of product/organizations (e.g., vehicles/dealerships) categories (e.g. Vehicle make—Dealer type (franchise, independent, etc. . . . ) can be evaluated as a function of relevant parameters {αn} (e.g., attributes). In order to compute the performance of the product category versus a market, in one example, machine-learning techniques can be utilized to extract the parameters {αn}, and a performance function g that depends on one or the extracted attributes can be computed. The value of the function g can be indicative or otherwise representative of the performance. In certain implementations, the function g can be the same as F. In one example, for the case N=9, the attributes {αn} that can be utilized to compute g can include one or more of the following, individually or in any combination:
α8: Total Stocked at auction; and
As illustrated and described herein, the content-based recommendation platform 110 can include a recommendation module 120 that can operate on a buying recommendation mode (e.g., buying mode 122) and selling recommendation mode (e.g., selling mode 132), and can leverage feature augmentation and meta-level hybridization techniques to generate a complementary set of vehicles and partner dealerships recommendations. In the buying mode 122 in accordance with this disclosure, in certain implementations, the recommendation module 120 can utilize or otherwise leverage as inputs the profile and model respectively generated by the profile learning module and business goal optimizer module, and can generate a ranked list of recommended products (e.g., vehicles) for each organization (e.g., a car dealership). Such a list of recommended products (e.g., recommended vehicles) can be utilized (e.g., hybridization) to generate a ranked list of recommended seller partner organizations (e.g., car dealerships) for each buying organization (e.g., a car dealership). In the selling mode 132 in accordance with this disclosure, in certain implementations, the recommendation module 120 can utilize or otherwise leverage at least a portion of the ranked list of recommended products (e.g., recommended vehicle(s) 128) and partners for each buyer dealership generated in buying mode to extract (e.g., hybridization) a ranked list of potential buyer partner dealerships for each vehicle of each seller (e.g., the selling dealership).
Example Functionality of the Content-Based Recommendation Platform 110 in Buying Mode.
Considering a buyer user (e.g., a car dealership), the recommendation module 120 can query the information storage 160 for all or at least some available vehicles that match one or more search criteria for a buying profile vector. In the present disclosure, a “profile vector” (either a buying profile vector or selling profile vector) can be embodied in or can include is a list of attributes associated with a user (either a buyer or a seller), each of the attributes representing a component of the profile vector. The attributes in the profile vector can be indicative or otherwise representative of the user's buying and/or selling behavior. For a user having a single profile vector, the “user profile” (e.g., buying profile or selling profile) can be represented by the single profile vector. In addition, for a user (buyer or seller) having multiple profile vectors, the user profile can be represented by the aggregation of the multiple profile vectors. For example, a user can have three profile vectors: (1) (2010, Honda, Accord), (2) (2011, Honda, Accord), and (3) (2005, Toyota, $8,000 to $9,000) (here, the third component is indicative of a typical (e.g., historical or preferred) price range). Thus, the profile of such a user is the aggregation of profile vectors (1) through (3). With regard to the query, in certain embodiments, the recommendation module 120 can query a database (relational or otherwise) or other data structures available in the information storage 160. The use of normalized data greatly reduces the computing requirements by allowing simple string or numerical matching between the buying profile vector and the vehicle attributes vector during the query. In addition or in the alternative, the matching can be realized by computing the cosine similarity between the two vectors. In certain embodiments, the query can also be performed using commercial full-text search engine (e.g., Solr/Lucene, ElasticSearch, or the like). Thus, in one example, this initial query can result in an unranked list of recommended vehicles.
In certain implementations, offer acceptance rating(s) can be determined or otherwise computed for each vehicle using the rating model(s) described herein in connection with maximizing offer acceptance rates. It should be appreciated that, in certain embodiments, the information (e.g., data and/or metadata) indicative of seller dealerships' selling profiles (e.g., selling profile(s) 148) can be considered during this computation, implicitly matching buyer's buying profiles (e.g., buying profile(s) 144) and seller's selling profiles (e.g., selling profile(s) 148). Such matching can result, in one aspect, in a rated and/or ranked list of recommended vehicles. Typically, this computation can be systematically performed for all or at least some of the “buyer” user (e.g., car dealerships) present in or otherwise accessible to the content-based recommendation platform 110 on a daily basis, according to another period (e.g., hourly, weekly), according to a schedule, and/or other type of timing protocol. Performing such a computation, either periodically or according to any schedule or timing protocol, can result in a rated and/or ranked list of recommended vehicles (e.g., recommended vehicle(s) 128) that can be stored or otherwise retained for subsequent hybridization operations in accordance with aspects of this disclosure.
In addition or in other implementations, the recommendation module 120 can utilize or otherwise leverage a rated and/or ranked list of recommended vehicles (e.g., recommended vehicle(s) 128) to generate a list of recommended seller partner dealerships, which can be referred to as “seller network” recommendations. As illustrated in
As illustrated in
Example Functionality of the Content-Based Recommendation Platform 110 in Selling Mode: Inventory Management.
Considering a seller user (e.g., a seller car dealership), the recommendation module is exploiting (e.g., hybridization) the rated and/or ranked list of recommended vehicles generated previously in buying mode. In one example, the recommendation module can determine or otherwise identify some or all the vehicles owned by the seller, which have been recommended to at least one buyer dealership(s), and append to these vehicle data their respective list of buyer dealerships, along with the previously computed vehicle ratings. In one example, these appended data are essentially a rated and/or ranked run list of buyers for each vehicle. The sum of these ratings is computed and stored as a “sellability” rating for each of the vehicles. The distribution of these ratings also can be retained for other uses (e.g., for implementing different sellability rating algorithms). Typically, this computation is systematically performed for all the “seller” user/dealerships in the system on daily basis, resulting in a rated and/or ranked list of recommended “vehicles-to-sell” that is stored for subsequent hybridization operations.
In addition to a rated and/or ranked list of recommended buyer(s) for each product (e.g., each vehicle), such as the recommended buyer(s) per vehicle 138, the buyer network recommendations for a seller user (e.g., seller dealership) can be extracted by aggregating the list of buyer dealerships appended to the list of recommended “vehicles-to-sell” into a list of unique buyers. To that end, in one example, the recommendation module 122, in selling mode, can determine or otherwise extract buyer network recommendations for a seller user by performing the aggregation described herein. In certain embodiments, the average of the previously computed vehicle ratings can be computed and stored as a buyer rating for each of these buyers. Alternatively or in additional embodiments, the rated and/or ranked list of recommended seller partner dealerships (e.g., recommended buyer network 134) generated previously in buying mode can be used to the same effect.
It should be appreciated that, in one aspect, the selling mode 132 as described herein of the recommendation module 120 can utilize or otherwise leverage as information input (which also may be referred to as “inputs”) the information (e.g., data and/or metadata) generated in buying mode 122. Accordingly, in certain implementations, the selling recommendations can be directed essentially to the products (e.g., vehicles) identified in buying mode. It should be appreciated that the latter illustrates the duality and complementarity of the recommendations generated in accordance with aspects of this disclosure. In certain embodiments, the recommendation module 120 can support or otherwise provide additional on-demand functionality generating a rated and/or ranked run list of buyer for an arbitrary product (e.g., arbitrary vehicle). In such a scenario, buying profile information (e.g., data and/or metadata, represented as buying profile(s) 144) generated by the profile learning module 140 of the content-based recommendation platform 110 can be directly queried to identify a list of unique buyers having buying profile vectors matching the arbitrary product. The sum of the weight of each of these vectors can be computed and utilized or otherwise leveraged as “buyer” rating for each of the buyers in the list of unique buyers.
It can be appreciated that, in certain embodiments, a generalized inventory management system encompassing all the vehicles in the dealership inventory also can be implemented. The generalized inventory management system can track (continuously or at certain frequency) the inventory of a seller (e.g., a car dealership), and can determine recommendations, based on the tracked inventory, for potential buyers; locations and/or times (e.g., time of the month) to sale a product (e.g., a vehicle) in the inventory; and/or price at which the product should be sold in order to maximize the financial and/or commercial interests of the seller.
As illustrated, the operator device 410 can include one or more input/output (I/O) interfaces 418 that can permit receiving and presenting information to an operator (or end-user). In certain embodiments, the I/O interface(s) 418 can include display units, including touch screens, reader devices (such as barcode scanners or NFC devices), haptic devices, network adapters, peripheral interfaces, or the like. One of the access module(s) 414 can receive at least some of the input information received via at least one of the I/O interface(s) 418. Similarly, information received by one or more of the access module(s) 414 from the content-based recommendation platform 110 can supplied to one or more of the I/O interface(s) 418 in order to present at least a portion of the information or a processed (e.g., rendered) version thereof to the operator. More specifically, in certain embodiments, one or more of the access module(s) 414 can permit generating and/or rendering information indicative or otherwise representative of user interfaces for consumption of content-based recommendations in accordance with aspects of this disclosure. In one example, a module of the access module(s) 414 can generate information indicative of selectable or otherwise actionable indicia that can be presented to an operator for providing information or otherwise interacting with the content-based recommendation platform 110. In addition or in other embodiments, such a module can implement the selection or actuation logic associated with the selectable or actionable indicia.
Certain indicia (selectable or otherwise) generated by at least one of the access module(s) 414 can convey a list of recommendation for a buyer and/or a seller. To that end, at least one of the I/O interface(s) 418 can present the indicia within a user interface. As an illustration,
With respect to example UI 600, the selling recommendations available or otherwise determined for a specific dealer (see, e.g., indicia 610) can include a recommended dealership, shown via indicia 620a and 620b, in the illustrated example selling recommendations. Each of the recommended dealerships, a group of selling recommendations for vehicles available via the recommended dealership. Each of the group of recommendations is ranked or otherwise rated via an offer acceptance rating (“likelihood of deal”), such a ranking shown with respective indicia 640a and 640b, for example. As described herein, each of the recommendations shown in the example UI 600 include information characterizing a recommended vehicle (or other type of product depending on the type of recommendation), such as “year make model trim;” “color;” “VIN;” “stock number;” “days in stock;” name, place of business, and type of the dealership associated with the dealer of recommended vehicles; and distance (“dist. (mi)”) from the dealership for which the selling recommendations are generated or otherwise determined (see, e.g., indicia 610). Further, the example UI 600 also can include selectable or otherwise actionable indicia 650 that, in response to selection or actuation, can cause a device presenting the UI 600 or another device coupled thereto to initiate and/or to send an electronic communication (e.g., an email) to the dealer associated with the recommended vehicle(s). For instance, actuation of the indicia 650 can cause the device to present a new window (e.g. a pop-up window) with various fillable fields or other types of indicia related to transmitting an email or another type of communication (e.g., a text message) to the dealer.
Example Applications of Weighted, Switching and Mixed Hybridization Techniques.
In certain embodiments, the profile learning module 140 and/or the business goal optimization module 150 can support multiple profile types and/or business goal optimization models. Accordingly, in one aspect, the content-based recommendation platform 110 can provide information (e.g., input data, input metadata, and/or instructions) to the recommendation module 120 for any practical combinations of these profiles and models. Examples of such combinations and associated recommendations can include the following: (I) Default Search Result Page. When a user/dealership initially logs into an inventory trading/direct sales website or platform and/or do not request specific search filters, the Search Result Page (SRP) is populated with a default list of vehicles. This list is preferentially generated using Weighted Hybridization, combining all the recommended vehicles identified for every buying profiles vectors of that user/dealership. The ratings from the different recommendation components are combined numerically for each vehicle, resulting in a rated and/or ranked search result page recommended vehicle list.
(II) Personalized Search Result Page. Alternatively or in other embodiments, a user (e.g., car dealership) can be given the capability to select the specific buying profiles (e.g., switching hybridization) to be used for a rated and/or ranked search result page presenting a recommended vehicle list or any other list of recommended products. As such, the search result page is customized (or personalized) to the selected buying profile.
(III) Account Management and Inventory Management Services. In certain embodiments, mixed hybridization can be used to display, simultaneously or otherwise, the recommendations for different combinations of (a) profiles (e.g., buying profile(s) and/or selling profile(s)) and (b) business goal optimization models, thus enabling account management and inventory management services.
(IV) Management of special sale events—e.g., Event Sales and “Ready to Move” sales). In certain implementations, a rated and/or ranked list of recommended vehicles generated in buying mode can be exploited to identify the run list of buyers for every vehicle marked as part of an event sale (e.g., sales occurring at specific predetermined times) of an inventory trading and/or direct sales website. In addition or in other implementations, the rated and/or ranked list of recommended vehicles can be utilized or otherwise leveraged to identify the run list of buyer for every vehicle marked or otherwise identified as being available to be traded under a particular condition (e.g., “ready to move” products). As such, a list of recommended products in accordance with the present disclosure can permit streamlining the reach-out to buyer strategy.
(V) Application to targeted prospecting. In certain implementations, a rated and/or ranked list of recommended vehicles generated in buying mode can be exploited to identify or otherwise determine a prospect list of dealerships that are not a member of an inventory trading website and/or a direct sales website, but have a large ratio of highly rated (or highly ranked) vehicles in their list of recommended vehicles and/or own vehicles that are highly rated (or highly ranked) in the list of recommended vehicles of an inventory trading and/or direct sales website's member. As such, list of recommended products in accordance with the present disclosure can permit thus streamlining the reach-out to prospect strategy. It should be appreciated that while the specific magnitude of high rating (or high rank) can be application and data distribution specific—e.g., 0.7 can be considered to be high in one use case, but average in another use case—, ratings equal to and/or higher than 0.80 can be typically considered as high ratings. It should further be appreciated that, in one example, a streamlined prospect strategy can permit managing or otherwise configuring resources to promote a trading platform associated with a content-based recommendation platform (e.g., platform 110) in accordance with one or more aspects of this disclosure.
Trading recommendations generated with a content-based recommendation platform in accordance with aspects of the disclosure can be utilized or otherwise leveraged for churn management (e.g., retention management) of organization that can trade in a trading platform associated with the organization that administers the content-based recommendation platform. In certain implementations, based on behavioral information (e.g., buying and/or selling behavior information) of an organization, the recommendation module can determine the likelihood of cancellation of the organization's subscription to the trading platform. Moreover, the recommendation platform can generate a trading recommendation directed to minimizing or otherwise reducing the likelihood of a subscription cancellation. In addition or in the alternative, by refining the quality and/or quantity of recommendations provided to the organization, the perceived quality of service of the content-based recommendation platform can increase, with the ensuing reduction of subscription cancellations at the trading platform.
The computational environment 700 illustrates an example implementation of the various aspects or features of the disclosure in which the processing or execution of operations described in connection with content-based trading recommendations disclosed herein can be performed at least in response to execution of one or more software components at the computing device 710. It should be appreciated that the one or more software components can render the computing device 710, or any other computing device that contains such components, a particular machine for providing content-based trading recommendations as described herein, among other functional purposes. A software component can be embodied in or can comprise one or more computer-accessible instructions, e.g., computer-readable and/or computer-executable instructions. In one scenario, at least a portion of the computer-accessible instructions can embody and/or can be executed to perform at least a part of one or more of the example methods described herein. For instance, to embody one such method, at least a portion of the computer-accessible instructions can be persisted (e.g., stored, made available, or stored and made available) in a computer storage non-transitory medium and executed by a processor. The one or more computer-accessible instructions that embody a software component can be assembled into one or more program modules that can be compiled, linked, and/or executed at the computing device 710 or other computing devices. Generally, such program modules comprise computer code, routines, programs, objects, components, information structures (e.g., data structures and/or metadata structures), etc., that can perform particular tasks (e.g., one or more operations) in response to execution by one or more processors, which can be integrated into the computing device 710 or functionally coupled thereto.
The various example embodiments of the disclosure can be operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that can be suitable for implementation of various aspects or features of the disclosure in connection with the cellular-sharing connectivity service described herein can comprise personal computers; server computers; laptop devices; handheld computing devices, such as mobile tablets and/or telephones; wearable computing devices; and multiprocessor systems. Additional examples can include set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, blade computers, programmable logic controllers (PLCs), distributed computing environments that comprise any of the above systems or devices, or the like.
As illustrated, the computing device 710 can comprise one or more processors 714, one or more input/output (I/O) interfaces 716, one or more memory devices 730 (herein referred to generically as memory 730), and a bus architecture 732 (also termed bus 732) that functionally couples various functional elements of the computing device 710. In certain embodiments, the computing device 710 can include, optionally, a radio unit 712. In such embodiments, the computing device 710 can embody or can constitute a consumption device or an injection device that operates wirelessly. The radio unit 712 can include one or more antennas and a communication processing unit (not shown in
The I/O interface(s) 716 can permit communication of information between the computing device and an external device, such as another computing device, e.g., a network element or an end-user device. Such communication can include direct communication or indirect communication, such as exchange of information between the computing device 710 and the external device via a network or elements thereof. As illustrated, the I/O interface(s) 716 can comprise one or more of network adapter(s) 718, peripheral adapter(s) 722, and rendering unit(s) 726. Such adapter(s) can permit or facilitate connectivity between the external device and one or more of the processor(s) 714 or the memory 730. For example, the peripheral adapter(s) 722 can include a group of ports, which can comprise at least one of parallel ports, serial ports, Ethernet ports, V.35 ports, or X.21 ports, wherein parallel ports can comprise one or more of GPIB ports and/or IEEE-1284 ports, while serial ports can include RS-232 ports, V.11 ports, USB ports, or FireWire or IEEE-1394 ports.
In one aspect, at least one of the network adapter(s) 718 can functionally couple the computing device 710 to one or more computing devices 770 via one or more traffic and signaling pipes 760 that can permit or facilitate exchange of traffic 762 and signaling 764 between the computing device 710 and the one or more computing devices 770. Such network coupling provided at least in part by the at least one of the network adapter(s) 718 can be implemented in a wired environment, a wireless environment, or a combination of both. The information that is communicated by the at least one of the network adapter(s) 718 can result from implementation of one or more operations in a method of the disclosure. Such output can include any form of visual representation, including, but not limited to, textual, graphical, animation, audio, tactile, or the like. In certain scenarios, each of the computing device(s) 770 can have substantially the same architecture as the computing device 710. In addition, or in the alternative, the rendering unit(s) 726 can include functional elements (e.g., lights, such as light-emitting diodes; a display, such as a LCD, a plasma monitor, a LED monitor, an electrochromic monitor; combinations thereof; or the like) that can permit control of the operation of the computing device 710, or can permit conveying or revealing the operational conditions of the computing device 710.
In one aspect, the bus 732 represents one or more of several possible types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. As an illustration, such architectures can comprise an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, an Accelerated Graphics Port (AGP) bus, and a Peripheral Component Interconnects (PCI) bus, a PCI-Express bus, a Personal Computer Memory Card Industry Association (PCMCIA) bus, a Universal Serial Bus (USB), and the like. The bus 732, and all buses described herein, can be implemented over a wired or wireless network connection, and each of the subsystems, including the processor(s) 714, the memory 730 and memory elements therein, and the I/O interface(s) 716 can be contained within one or more remote computing devices 770 at physically separate locations, connected through buses of this form, thereby effectively implementing a fully distributed system. In distributed systems, the content-recommendation component(s) can be distributed amongst the computing device 710 and the remote computing devices 770.
The computing device 710 can comprise a variety of computer-readable media. Computer-readable media can be any available media (transitory and non-transitory) that can be accessed by a computing device. In one aspect, computer-readable media can comprise computer non-transitory storage media (or computer-readable non-transitory storage media) and communications media. Example computer-readable non-transitory storage media can be any available media that can be accessed by the computing device 710, and can comprise, for example, both volatile and non-volatile media, and removable and/or non-removable media. In one aspect, the memory 730 can comprise computer-readable media in the form of volatile memory, such as random access memory (RAM), and/or non-volatile memory, such as read-only memory (ROM).
The memory 730 can comprise functionality instructions storage 734 and functionality information storage 738. The functionality instructions storage 734 can comprise computer-accessible instructions that, in response to execution by at least one of the processor(s) 714, can implement one or more of the functionalities of the disclosure. The computer-accessible instructions can embody or can comprise one or more software components illustrated as content-based recommendation component(s) 736. In one scenario, execution of at least one component of the content-based recommendation component(s) 736 can implement at least a portion of the functionality described in the present disclosure, and/or one or more of the methods described herein, such as example methods 1000 and 1100. For instance, such execution can cause a processor that executes the at least one component to carry out a disclosed example method. It should be appreciated that, in one aspect, a processor of the processor(s) 714 that executes at least one of the content-based recommendation component(s) 736 can retrieve information from or retain information in a memory element 740 in the functionality information storage 738 in order to operate in accordance with the functionality programmed or otherwise configured by the content-based recommendation component(s) 736. Such information can include buying profiles, selling profiles, rating models, product recommendations, a combination thereof, or the like. In addition, such information can include at least one of programming code instructions (or code instructions), information structures, or the like. At least one of the one or more interfaces 750 (e.g., application programming interface(s)) can permit or facilitate communication of information between two or more components within the functionality instructions storage 734. The information that is communicated by the at least one interface can result from implementation of one or more operations in a method of the disclosure or any other functionality described herein. In certain embodiments, one or more of the functionality instructions storage 734 and the functionality information storage 738 can be embodied in or can comprise removable/non-removable, and/or volatile/non-volatile computer storage media.
At least a portion of at least one of the content-based recommendation component(s) 736 or content-based recommendation information 740 can program or otherwise configure one or more of the processors 714 to operate at least in accordance with the functionality described in the present disclosure. In one embodiment, the content-based recommendation component(s) 736 contained in the functionality instruction(s) storage 734 can include the profile learning module in content-based recommendation platform 110, the business goal optimization module 150 in the content-based recommendation platform 110 shown in
With further reference to
In addition, the memory 730 can comprise computer-accessible instructions and information (e.g., data, metadata, and/or programming code) that permit or otherwise facilitate operation and/or administration (e.g., upgrades, software installation, any other configuration, or the like) of the computing device 710. Accordingly, as illustrated, the memory 730 can comprise a memory element 742 (labeled operating system (OS) instruction(s) 742) that can contain one or more program modules that embody or include one or more operating systems, such as a Windows operating system, Unix, Linux, Symbian, Android, Chromium, or substantially any operating system suitable for mobile computing devices or tethered computing devices. In one aspect, the operational and/or architectural complexity of the computing device 710 can dictate a suitable operating system. The memory 730 also comprises a system information storage 746 having data and/or metadata that permits or facilitates operation and/or administration of the computing device 710. Elements of the OS instruction(s) 742 and the system information storage 746 can be accessible or can be operated on by at least one of the processor(s) 714.
It should be recognized that while the functionality instructions storage 734 and other executable program components, such as the OS instruction(s) 742, are illustrated herein as discrete blocks, such software components can reside at various times in different memory components of the computing device 710, and can be executed by at least one of the processor(s) 714. In certain scenarios, an implementation of the content-based recommendation component(s) 736 can be retained on or transmitted across some form of computer-readable media.
The computing device 710 and/or one of the computing device(s) 770 can include a power supply (not shown), which can power up components or functional elements within such devices. The power supply can be a rechargeable power supply, e.g., a rechargeable battery, and it can include one or more transformers to achieve a power level suitable for operation of the computing device 710 and/or one of the computing device(s) 770, and components, functional elements, and related circuitry therein. In certain scenarios, the power supply can be attached to a conventional power grid to recharge and ensure that such devices can be operational. In one aspect, the power supply can include an I/O interface (e.g., one of the network adapter(s) 718) to connect operationally to the conventional power grid. In another aspect, the power supply can include an energy conversion component, such as a solar panel, to provide additional or alternative power resources or autonomy for the computing device 710 and/or at least one of the computing device(s) 770.
The computing device 710 can operate in a networked environment by utilizing connections to one or more remote computing devices 770. As an illustration, a remote computing device can be a personal computer, a portable computer, a server, a router, a network computer, a peer device or other common network node, and so on. As described herein, connections (physical and/or logical) between the computing device 710 and a computing device of the one or more remote computing devices 770 can be made via one or more traffic and signaling pipes 760, which can comprise wireline link(s) and/or wireless link(s) and several network elements (such as routers or switches, concentrators, servers, and the like) that form a local area network (LAN), a metropolitan area network (MAN), and/or a wide area network (WAN). Such networking environments are conventional and commonplace in dwellings, offices, enterprise-wide computer networks, intranets, local area networks, and wide area networks.
In one or more embodiments, one or more of the disclosed methods can be practiced in distributed computing environments, such as grid-based environments, where tasks can be performed by remote processing devices (computing device(s) 770) that are functionally coupled (e.g., communicatively linked or otherwise coupled) through a network having traffic and signaling pipes and related network elements. In a distributed computing environment, in one aspect, one or more software components (such as program modules) can be located in both a local computing device (e.g. the computing device 710) and at least one remote computing device.
In view of the aspects described herein, example methods that can be implemented in accordance with the present disclosure can be better appreciated with reference to the flowcharts in
It should be appreciated that the methods of the disclosure can be retained on an article of manufacture, or computer-readable medium, to permit or facilitate transporting and transferring such methods to a computing device (e.g., a desktop computer; a mobile computer, such as a tablet, or a smartphone; a gaming console; a mobile telephone; a blade computer; a programmable logic controller; and the like) for execution and thus, implementation by a processor of the computing device or for storage in a memory device (or memory) thereof or functionally coupled thereto. In one aspect, one or more processors, such as processor(s) that implement (e.g., compile, link, and/or execute) one or more of the disclosed methods, can be employed to execute instructions (e.g., programming instructions) retained in a memory, or any computer- or machine-readable medium, to implement at least one of the one or more methods. The instructions can provide a computer-executable or machine-executable framework to implement the methods disclosed herein.
Various embodiments of the disclosure may take the form of an entirely or partially hardware embodiment, an entirely or partially software embodiment, or a combination of software and hardware (e.g., a firmware embodiment). Furthermore, as described herein, various embodiments of the disclosure (e.g., methods and systems) may take the form of a computer program product comprising a computer-readable non-transitory storage medium having computer-accessible instructions (e.g., computer-readable and/or computer-executable instructions) such as computer software, encoded or otherwise embodied in such storage medium. Those instructions can be read or otherwise accessed and executed by one or more processors to perform or permit performance of the operations described herein. The instructions can be provided in any suitable form, such as source code, compiled code, interpreted code, executable code, static code, dynamic code, assembler code, combinations of the foregoing, and the like. Any suitable computer-readable non-transitory storage medium may be utilized to form the computer program product. For instance, the computer-readable medium may include any tangible non-transitory medium for storing information in a form readable or otherwise accessible by one or more computers or processor(s) functionally coupled thereto. Non-transitory storage media can include read only memory (ROM); random access memory (RAM); magnetic disk storage media; optical storage media; flash memory, etc.
Embodiments of the operational environments and methods (or techniques) are described herein with reference to block diagrams and flowchart illustrations of methods, systems, apparatuses and computer program products. It can be understood that each block of the block diagrams and flowchart illustrations, and combinations of blocks in the block diagrams and flowchart illustrations, respectively, can be implemented by computer-accessible instructions. In certain implementations, the computer-accessible instructions may be loaded or otherwise incorporated into onto a general purpose computer, special purpose computer, or other programmable information processing apparatus to produce a particular machine, such that the operations or functions specified in the flowchart block or blocks can be implemented in response to execution at the computer or processing apparatus.
Unless otherwise expressly stated, it is in no way intended that any protocol, procedure, process, or method set forth herein be construed as requiring that its acts or steps be performed in a specific order. Accordingly, where a process or method claim does not actually recite an order to be followed by its acts or steps or it is not otherwise specifically recited in the claims or descriptions of the subject disclosure that the steps are to be limited to a specific order, it is in no way intended that an order be inferred, in any respect. This holds for any possible non-express basis for interpretation, including: matters of logic with respect to arrangement of steps or operational flow; plain meaning derived from grammatical organization or punctuation; the number or type of embodiments described in the specification or annexed drawings, or the like.
As used in this application, the terms “component,” “environment,” “system,” “platform,” “architecture,” “interface,” “unit,” “pipe,” “module,” “source,” and the like are intended to refer to a computer-related entity or an entity related to an operational apparatus with one or more specific functionalities. Such entities may be either hardware, a combination of hardware and software, software, or software in execution. As an example, a component may be, but is not limited to being, a process running on a processor, a processor, an object, an executable portion of software, a thread of execution, a program, and/or a computing device. For example, both a software application executing on a computing device and the computing device can be a component. One or more components may reside within a process and/or thread of execution. A component may be localized on one computing device or distributed between two or more computing devices. As described herein, a component can execute from various computer-readable non-transitory media having various data structures stored thereon. Components can communicate via local and/or remote processes in accordance, for example, with a signal (either analogic or digital) having one or more data packets (e.g., data from one component interacting with another component in a local system, distributed system, and/or across a network such as a wide area network with other systems via the signal). As another example, a component can be an apparatus with specific functionality provided by mechanical parts operated by electric or electronic circuitry that is controlled by a software application or firmware application executed by a processor, wherein the processor can be internal or external to the apparatus and can execute at least a part of the software or firmware application. As yet another example, a component can be an apparatus that provides specific functionality through electronic components without mechanical parts, the electronic components can include a processor therein to execute software or firmware that provides, at least in part, the functionality of the electronic components. An interface can include input/output (I/O) components as well as associated processor, application, and/or other programming components. The terms “component,” “environment,” “system,” “platform,” “architecture,” “interface,” “unit,” “pipe,” and “module” can be utilized interchangeably and can be referred to collectively as functional elements.
In the present specification and annexed drawings, reference to a “processor” is made. As utilized herein, a processor can refer to any computing processing unit or device comprising single-core processors; single-processors with software multithread execution capability; multi-core processors; multi-core processors with software multithread execution capability; multi-core processors with hardware multithread technology; parallel platforms; and parallel platforms with distributed shared memory. Additionally, a processor can refer to an integrated circuit (IC), an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic controller (PLC), a complex programmable logic device (CPLD), a discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor can be implemented as a combination of computing processing units. In certain embodiments, processors can utilize nanoscale architectures such as, but not limited to, molecular and quantum-dot based transistors, switches and gates, in order to optimize space usage or enhance the performance of user equipment or other electronic equipment.
In addition, in the present specification and annexed drawings, terms such as “store,” storage,” “data store,” “data storage,” “memory,” “repository,” and substantially any other information storage component relevant to operation and functionality of a component of the disclosure, refer to “memory components,” entities embodied in a “memory,” or components forming the memory. It can be appreciated that the memory components or memories described herein embody or comprise non-transitory computer storage media that can be readable or otherwise accessible by a computing device. Such media can be implemented in any methods or technology for storage of information such as computer-readable instructions, information structures, program modules, or other information objects. The memory components or memories can be either volatile memory or non-volatile memory, or can include both volatile and non-volatile memory. In addition, the memory components or memories can be removable or non-removable, and/or internal or external to a computing device or component. Example of various types of non-transitory storage media can comprise hard-disc drives, zip drives, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, flash memory cards or other types of memory cards, cartridges, or any other non-transitory medium suitable to retain the desired information and which can be accessed by a computing device.
As an illustration, non-volatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM), which acts as external cache memory. By way of illustration and not limitation, RAM is available in many forms such as synchronous RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The disclosed memory components or memories of operational environments described herein are intended to comprise one or more of these and/or any other suitable types of memory.
Conditional language, such as, among others, “can,” “could,” “might,” or “may,” unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain implementations could include, while other implementations do not include, certain features, elements, and/or operations. Thus, such conditional language generally is not intended to imply that features, elements, and/or operations are in any way required for one or more implementations or that one or more implementations necessarily include logic for deciding, with or without user input or prompting, whether these features, elements, and/or operations are included or are to be performed in any particular implementation.
What has been described herein in the present specification and annexed drawings includes examples of systems, devices, and techniques that can provide content-based recommendations for organization-to-organization trading. As an illustration, a flexible and scalable content-based hybrid recommendation platform and techniques has been described, where such a platform and/or techniques permit generation of a complementary set of dual recommendations for vehicles and dealers networks from a buying and selling perspective. In addition, in certain implementations, the disclosure can leverage or otherwise utilize an optimization module and hybridization techniques to address, in at least certain aspects, various business operational scenarios of an organization (e.g., a car dealership or the administrator of the recommendation platform). It is, of course, not possible to describe every conceivable combination of elements and/or methods for purposes of describing the various features of the disclosure, but it can be recognized that many further combinations and permutations of the disclosed features are possible. Accordingly, it may be apparent that various modifications can be made to the disclosure without departing from the scope or spirit thereof. In addition or in the alternative, other embodiments of the disclosure may be apparent from consideration of the specification and annexed drawings, and practice of the disclosure as presented herein. It is intended that the examples put forward in the specification and annexed drawings be considered, in all respects, as illustrative and not restrictive. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.
This application relates to and claims priority from U.S. Provisional Application No. 61/926,237, filed Jan. 10, 2014, which is hereby incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61926237 | Jan 2014 | US |
