Patent Grant
9460214
This application is related to U.S.application Ser. No. 13/730,765, filed Dec. 28, 2012 and U.S. application Ser. No. 13/730,781, filed Dec. 28, 2012. The application is incorporated herein by reference for all purposes.
The present disclosure relates to systems and methods that rank search results based on a reference to a color in a search query.
Purchasers of products or services may perform various search queries to locate a desired product or service. In some situations, the purchaser may include one or more characteristics of the desired product or service, such as color, size, product type, and the like. However, many existing search engines do not properly select products or services that have the desired characteristics. Instead, at least a portion of the search results displayed to the purchaser are not what the purchaser desired. Further, many existing search engines do not properly rank the search results based on the desired characteristics provided by the purchaser.
Non-limiting and non-exhaustive embodiments of the present disclosure are described with reference to the following figures, wherein like reference numerals refer to like parts throughout the various figures unless otherwise specified.
In the following description, reference is made to the accompanying drawings that form a part thereof, and in which is shown by way of illustration specific exemplary embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the concepts disclosed herein, and it is to be understood that modifications to the various disclosed embodiments may be made, and other embodiments may be utilized, without departing from the scope of the present disclosure. The following detailed description is, therefore, not to be taken in a limiting sense.
Reference throughout this specification to “one embodiment,” “an embodiment,” “one example,” or “an example” means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present disclosure. Thus, appearances of the phrases “in one embodiment,” “in an embodiment,” “one example,” or “an example” in various places throughout this specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, databases, or characteristics may be combined in any suitable combinations and/or sub-combinations in one or more embodiments or examples. In addition, it should be appreciated that the figures provided herewith are for explanation purposes to persons ordinarily skilled in the art and that the drawings are not necessarily drawn to scale.
Embodiments in accordance with the present disclosure may be embodied as an apparatus, method, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware-comprised embodiment, an entirely software-comprised embodiment (including firmware, resident software, micro-code, etc.), or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “circuit,” “module,” or “system.” Furthermore, embodiments of the present disclosure may take the form of a computer program product embodied in any tangible medium of expression having computer-usable program code embodied in the medium.
Any combination of one or more computer-usable or computer-readable media may be utilized. For example, a computer-readable medium may include one or more of a portable computer diskette, a hard disk, a random access memory (RAM) device, a read-only memory (ROM) device, an erasable programmable read-only memory (EPROM or Flash memory) device, a portable compact disc read-only memory (CDROM), an optical storage device, and a magnetic storage device. Computer program code for carrying out operations of the present disclosure may be written in any combination of one or more programming languages. Such code may be compiled from source code to computer-readable assembly language or machine code suitable for the device or computer on which the code will be executed.
Embodiments may also be implemented in cloud computing environments. In this description and the following claims, “cloud computing” may be defined as a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned via virtualization and released with minimal management effort or service provider interaction and then scaled accordingly. A cloud model can be composed of various characteristics (e.g., on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service), service models (e.g., Software as a Service (“SaaS”), Platform as a Service (“PaaS”), and Infrastructure as a Service (“IaaS”)), and deployment models (e.g., private cloud, community cloud, public cloud, and hybrid cloud).
The flow diagrams and block diagrams in the attached figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It will also be noted that each block of the block diagrams and/or flow diagrams, and combinations of blocks in the block diagrams and/or flow diagrams, may be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. These computer program instructions may also be stored in a computer-readable medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable medium produce an article of manufacture including instruction means which implement the function/act specified in the flow diagram and/or block diagram block or blocks.
The systems and methods described herein generate search results based on references to one or more colors in a search query submitted by a user. As described herein, a product search query is received from a user. The product search query is analyzed to identify a reference to a color in the product search query. The search query is executed to identify multiple products arranged in an initial ranking order. A title and/or description of each of the multiple products are analyzed to identify references to color. A modified ranking of the multiple products is created by increasing the raking of products that include a reference to the color in the title and/or description.
Online marketplace 102 communicates with various systems, services, and devices through data communication network 104. Data communication network 104 may utilize any communication protocol and any type of communication medium. In some embodiments, data communication network 104 is a combination of two or more networks coupled to one another. Online marketplace 102 also communicates with various systems and devices, such as mobile devices, through cellular communication network 106, which may utilize any communication protocol and any type of communication medium. In some embodiments, cellular communication network 106 is a combination of two or more networks coupled to one another.
As shown in environment 100, a mobile device 110 communicates with online marketplace 102 through cellular communication network 106. Although a single mobile device 110 is shown in
Additionally, a user device 114 communicates with online marketplace 102 through data communication network 104. User device 114 includes any type of device capable of communicating with online marketplace 102 through data communication network 104, such as a smart phone, a tablet computer, a laptop computer, a desktop computer, a portable entertainment device, a portable gaming device, a game console, a set top box, and the like.
An optional color analysis service 116 is also coupled to data communication network 104. Color analysis service 116 performs various functions related to the identification and analysis of colors in a search query, as discussed herein. For example, color analysis service 116 may analyze a particular color referenced in a search query to identify similar colors, such as identifying the color “pink” as being similar to the color “red.” Although a single color analysis service 116 is shown in
Server 200 also includes product search module 208, which handles the receiving and processing of product search queries (or product search requests) from multiple users. In some embodiments, product search module 208 supports various product search techniques, such as a keyword search, a category search, a department search, a brand search, and the like. In particular implementations, users interact with server 200 through a user interface to search for desired products or services. A search query processing module 210 performs, for example, various operations associated with the processing of product search queries received from multiple users. As discussed herein, search query processing module 210 may execute a search query to identify potential products or services of interest to the user. In some embodiments, execution of the search query identifies multiple products that are initially ranked based on a likelihood of interest to the user providing the search query (e.g., products with a highest likelihood of interest to the user are presented at the top of the ranking).
A color analysis module 212 performs various functions associated with identifying references to one or more colors in a search query. As discussed herein, any reference to a color in a search query may affect the ranking of the search results presented to the user. Thus, color analysis module 212 may communicate with search query processing module 210 and a search result ranking module 214 to assist with the creation of appropriately ranked search results. In some embodiments, search result ranking module 214 may determine an initial ranking based on an initial execution of the search query submitted by the user. As discussed herein, search results ranking module 214 may then modify the initial ranking of the search results based on references to color in the search query and other factors.
Server 200 further includes a user interface generator 216 that creates data to present various user interfaces to a user of mobile device 110, 112, or user device 114. Example user interfaces include search query input interfaces, product or service listings based on the results of executing the search query, and the like. A data communication bus 218 allows the various systems and components of server 200 to communicate with one another.
User device 114 also includes a display generator 308, which generates various signals that enable a display device to present information to a user of the device. In some embodiments, display generator 308 generates various signals that present a user interface to the user of user device 114. In particular implementations, display generator 308 includes, or is coupled to, a display device to present information to the user of the device. This user interface allows a user to, for example, submit a product search query to online marketplace 102. A user input device 310 allows a user to interact with user device 114. Example user input devices 310 include pointing devices, buttons, switches, touch-sensitive portions of a touch-sensitive display device, and the like. A data communication bus 312 allows the various systems and components of user device 114 to communicate with one another. In some embodiments, systems and components similar to those discussed above with respect to user device 114 are included in mobile devices 110 and 112.
Initially, the method 400 accesses a search query generated by a user at 402. As mentioned above, method 400 may be applied to any type of search query (or combination of multiple types of search queries). For this example, the search query includes at least one keyword entered by the user. The search query may be accessed from another system or may be received directly from the user. The method 400 analyzes the search query to identify any references to a color in the search query at 404. For example, a search query “brown leather jacket” includes a reference to the color “brown,” whereas a search query “small leather jacket” does not include any reference to color. Some search queries may contain references to multiple colors, such as “red or pink purse.” In some embodiments, each word or term in a search query is compared against a database (or other data listing) of known color terms to identify references to a color in the search query.
If a color is identified in the search query, the method 400 analyzes the search query to determine whether the user intended an actual color in the search query at 406. In some situations, a user may include a term in a search query that matches a known color term, but the user did not intend the term to identify an actual color. For example, a search query for “Red Sox shirt” is referring to the baseball team “Red Sox”, not the color red. In this example, the user did not intend an actual color, so the term “red” is not considered in modifying the rankings of the search results. Referring to
However, if an actual color is intended at 408, the method 400 identifies multiple products associated with the search query at 412 and creates an initial ranking of the multiple products associated with the search query at 414. A color associated with a particular product is determined, for example, based on a title of the product, a description of the product, a color parameter associated with the product, and the like. For each identified product, the method 400 extracts references to the color in the title and description of the identified product at 416.
Referring to
The method 400 continues by determining an initial ranking of the multiple products identified using the search query at 422. An updated ranking of the multiple products is generated at 424 based on colors associated with the multiple products and the color similarity score associated with the multiple products. For example, consider a situation where two products initially have a similar ranking based on a search query that includes a color term. If one of the products has a color identical to the color term in the search query and the other product has a different color (or dissimilar color), the product with the identical color will be ranked ahead of the other product. Additionally, if both products have colors that are similar to the color term in the search query, the product with the highest color similarity will be ranked ahead of the other product. The method 400 concludes by displaying the updated ranking of the multiple products to the user at 426. In alternate embodiments, the method 400 generates data that is used by another device or system to present the updated ranking of the multiple products to the user. In some embodiments, a score is assigned to each of the multiple products and the ranking of the multiple products is performed based on the assigned score. In a particular implementation, the BM25F ranking function is used to rank the multiple products.
In alternate embodiments, the method 400 is performed without identifying or analyzing additional colors that are similar to the color in the identified search query. In these alternate embodiments, the method 400 does not consider color similarity when ranking the search results. Instead, the method 400 considers exact color matches between colors identified in the user search query and colors associated with the multiple products associated with the search query when ranking the multiple products.
As noted above, a color associated with a particular product may be determined based on a title of the product, a description of a product, a color parameter associated with the product, and the like. In some embodiments, products available in an online marketplace are periodically analyzed to identify color attributes associated with the products. For example, if a particular product does not have a color associated with a “color field” describing the product, a procedure analyzes all data associated with the product to determine one or more colors associated with the product. For example, the title, description, and features of a product are analyzed to identify actual colors associated with the product. If one or more colors are identified, those colors are added to a “color field” related to that product. In some embodiments, the title, description, and features of a product are analyzed using the systems and methods described herein to identify actual colors associated with the product.
The resulting value of the cosine similarity calculation mentioned above determines whether the user intended an actual color in the search query. In one embodiment, a cosine similarity of 0.8 is used as a threshold for determining whether the user intended an actual color in the search query. In this embodiment, if the cosine similarity is greater than or equal to 0.8 at 512, the method 500 presumes that the user did not intend an actual color in the search query. In this situation, the method 500 executes the search query and ranks the search results without regard to color at 514. However, if the cosine similarity is less than 0.8 at 512, the method 500 presumes that the user intended an actual color in the search query. In this situation, the method 500 applies the identified color (or colors) in the search query to modify the search results ranking at 516.
After determining a color similarity between the color term in the search query and each of the known color terms, the method 600 determines whether any pairs of colors are similar at 608. If none of the known color terms are similar to the color term in the search query, the method 600 generates a message indicating the lack of similar colors at 610. This message is communicated to a system or procedure that is handling the ranking of the search results. If, at 608, at least one of the known color terms are similar to the color term in the search query, the method 600 creates a listing of similar colors and a color similarity score for each color in the listing at 612. This listing is used by the system or procedure that is handling the ranking of the search results, as discussed herein.
In a particular example, a color term “red” is used in a search query. Several known colors are determined to have at least a minimal level of color similarity, such as “pink,” “burgundy,” and “maroon.” In this example, a color similarity score is calculated for each of the (red-pink), (red-burgundy), and (red-maroon) pairs. The following table illustrates example color similarity scores for the three identified pairs. In this example, a lower similarity score represents a higher similarity between the two colors.
In this example, the color pink has the highest similarity with the “red” color term in the search query. Thus, a product having a pink color is given a higher rank in the search results when compared to products having burgundy or maroon colors. Additionally, a product having a burgundy or maroon color is given a higher rank in the search results when compared to products having dissimilar colors, such as blue, green, silver, and yellow.
In a particular implementation, a color similarity score is calculated using a color difference formula published by the International Commission on Illumination (CIE), commonly referred to as CIE94. This formula determines a difference (or distance) between two colors.
The method 700 continues as the mobile device receives a user selection of a particular product presented to the user at 710. For example, the user may desire additional information regarding a particular product. The mobile device communicates the user selection to the online marketplace at 712 and receives data associated with the selected product from the online marketplace at 714. The mobile device presents the data associated with the selected product to the user at 716. The user may then request additional information about the product, request information related to a different product, place an order to purchase the product, generate a new search query, and the like.
Although the present disclosure is described in terms of certain preferred embodiments, other embodiments will be apparent to those of ordinary skill in the art, given the benefit of this disclosure, including embodiments that do not provide all of the benefits and features set forth herein, which are also within the scope of this disclosure. It is to be understood that other embodiments may be utilized, without departing from the scope of the present disclosure.
| Number | Name | Date | Kind |
|---|---|---|---|
| 5751450 | Robinson | May 1998 | A |
| 5949904 | Delp | Sep 1999 | A |
| 7519548 | Hanechak et al. | Apr 2009 | B2 |
| 7536059 | Xu et al. | May 2009 | B2 |
| 7657100 | Gokturk | Feb 2010 | B2 |
| 8249398 | Lee | Aug 2012 | B2 |
| 8406573 | Wang | Mar 2013 | B2 |
| 8576241 | Kanter | Nov 2013 | B1 |
| 8577134 | Tirumalareddy | Nov 2013 | B2 |
| 8611695 | Han | Dec 2013 | B1 |
| 8655648 | Kulkarni et al. | Feb 2014 | B2 |
| 8699826 | Remedios | Apr 2014 | B2 |
| 8712156 | Bronstein | Apr 2014 | B2 |
| 8891860 | Masuko | Nov 2014 | B2 |
| 20030217052 | Rubenczyk et al. | Nov 2003 | A1 |
| 20030231185 | Kupersmit | Dec 2003 | A1 |
| 20040101156 | Kacker | May 2004 | A1 |
| 20060047580 | Saha | Mar 2006 | A1 |
| 20070018906 | Visnovsky et al. | Jan 2007 | A1 |
| 20070223811 | Kudo | Sep 2007 | A1 |
| 20080046410 | Lieb | Feb 2008 | A1 |
| 20080046424 | Horton | Feb 2008 | A1 |
| 20080082518 | Loftesness | Apr 2008 | A1 |
| 20080089615 | Shiiyama | Apr 2008 | A1 |
| 20080118151 | Bouguet | May 2008 | A1 |
| 20080152225 | Iwamoto | Jun 2008 | A1 |
| 20080154608 | Evermann et al. | Jun 2008 | A1 |
| 20080154747 | Tarbell | Jun 2008 | A1 |
| 20080162469 | Terayoko | Jul 2008 | A1 |
| 20100049707 | Faria et al. | Feb 2010 | A1 |
| 20100158412 | Wang et al. | Jun 2010 | A1 |
| 20110191334 | Hua | Aug 2011 | A1 |
| 20110235902 | Chittar et al. | Sep 2011 | A1 |
| 20120185472 | Ahmed et al. | Jul 2012 | A1 |
| 20120296926 | Kalin et al. | Nov 2012 | A1 |
| 20130022264 | Atsmon | Jan 2013 | A1 |
| 20130080426 | Chen | Mar 2013 | A1 |
| Entry |
|---|
| Xiao et al. (“Extracting Structured Information from User Queries with Semi-Supervised Conditional Random Fields”, pp. 572-579, ACM New York, NY USA © 2009; ISBN:978-1-60558-483-6). |
| Number | Date | Country | |
|---|---|---|---|
| 20140188854 A1 | Jul 2014 | US |
