Systems and methods for generating security improvement plans for entities

Abstract

A computer-implemented method is provided for statistical modeling of entities of a particular type. The method can include obtaining entity data including a plurality of entity data sets, each entity data set associated with a respective entity and including values for one or more static parameters indicative of a type of the entity. Each entity data set can include (i) values for input parameter(s) indicative of a security profile of the entity and (ii) a value of a security class parameter indicative of a security class of the entity based on the values of the input parameters. The method can include training a statistical classifier to infer a value of the security class parameter indicative of the security class of a particular entity of the particular type based on values of one or more of the input parameters indicative of a security profile of the particular entity.

Description

TECHNICAL FIELD

The following disclosure is directed to methods and systems for generating a security improvement plan for an entity and, more specifically, methods and systems for generating a security improvement plan for an entity based on security ratings of similar entities.

BACKGROUND

Ratings enable quantitative comparisons among entities (e.g., companies, students, automobiles, etc.). For example, ratings can be used by consumers to determine whether to buy from a particular company. In another example, ratings can be used by potential employees to determine whether to work at particular company. Thus, entities subject to a ratings scheme typically strive to improve their respective ratings to enhance their standing in their industry or community. One type of ratings scheme pertains to the security of an entity. Specifically, an entity (e.g., a company) can be rated based on past cybersecurity events and/or future cybersecurity risks. Aside from the company itself, there may be multiple stakeholders, e.g., insurance companies, business partners, and clients, that are invested in an improved security rating of the particular company.

Conventional methods utilize brittle rules or to use summary statistics from a vast data set to derive improvement plans. However, these methods can lead to crude or unrealistic plans for most entities.

SUMMARY

Disclosed herein are systems and methods for generating a security improvement plan for an entity with the goal of improving its security rating. An entity can include an organization, a company, a group, a school, a government, etc. An entity may be characterized by one or more static parameters, e.g., entity size, entity industry, entity location, etc., as these aspects of the entity do not typically change. It is understood that some of these aspects may indeed change over time (e.g., an entity may sell off a part of its business resulting in a decreased size, or it may venture into a new industry or location). In some embodiments, if a static parameter of an entity changes, the statistical classifier, as discussed below, may be retrained based on the changed value of the static parameter.

An improved security rating reflects improvements made to the security profile of the entity. Specifically, the security profile of an entity can be indicated by one or more input parameters, e.g., a number of botnet infections of the entity's computer network or a number of malware-infected servers associated with the entity. These input parameters are typically modifiable in that an entity can change or improve the value of the parameter, thereby improving its security rating. For example, an entity can strive to decrease the number of botnet infections or decrease the number of malware-infected servers. By doing so, an entity's security rating may increase, e.g., from 680 to 720, indicating an improved ability to withstand or prevent cybersecurity attacks. An improved security rating can also increase confidence of various stakeholders of the entity that the entity is more secure and/or protected from cybersecurity risks that it had previously been.

In many instances, for a given entity, the number of input parameters can be significant and, therefore, the space of possible improvement plans can be quite large. In some embodiments, the respective values of two or more input parameters may be correlated with one another. In some embodiments, the respective values of two or more input parameters may be interdependent. Though these relationships may exist, in many instances, the relationships may not be apparent or well-understood. This suggests that some or much of an entity's improvement plan space may not be sensible or achievable. In some cases, some of these improvement plans are unachievable by the particular entity because they are dependent on parameters that are difficult for the particular entity to modify. Therefore, determining an achievable improvement plan for an entity can be difficult due to the large space of possible plans and the implausibility of many areas of the space. Therefore, generating an achievable security improvement plan can depend on reducing the large space of possible improvement plans and eliminating unachievable portions of improvement plans.

Some embodiments of systems and methods described herein are configured to generate a feasible security improvement plan for the entity. A feasible security improvement plan is important to provide the entity with realistic, achievable goals with a reasonable expectation and/or a reasonable likelihood of achieving those goals. A security improvement plan can include value(s) for one or more modifiable input parameters of the entity such that the value(s) contribute to an increase in the security rating of the entity. The exemplary systems and method described herein can focus the space of possible improvement plans by using data related to similar entities that share the parameters of a particular entity for which the plan is generated.

In accordance with an embodiment of the disclosure, a computer-implemented method is provided for statistical modeling of entities of a particular type. The method can include obtaining entity data including a plurality of entity data sets, each entity data set associated with a respective entity and including values for one or more static parameters indicative of a type of the entity. The values of the static parameters for each of the entity data sets can indicate that the type of the entity matches the particular type, and each entity data set can include (i) values for one or more input parameters indicative of a security profile of the entity and (ii) a value of a security class parameter indicative of a security class of the entity based on the values of the input parameters. The method can include training a statistical classifier to infer a value of the security class parameter indicative of the security class of a particular entity of the particular type based on values of one or more of the input parameters indicative of a security profile of the particular entity. The training the statistical classifier can include fitting the statistical classifier to the plurality of entity data sets.

Various embodiments of the method can include one or more of the following features. The static parameters can include (i) entity size, (ii) entity industry, and/or (iii) entity location. The values of two or more static parameters for each of the entity data sets can indicate that the type of the entity matches the particular type. The method can include selecting a target value for the security class parameter indicative of the security class for the particular entity. The plurality of entity data sets can include one or more entity data sets for which the value of the security class parameter is lower than the target value and one or more entity data sets for which the value of the security class parameter is at or above than the target value. The plurality of entity data sets includes at least three entity data sets for which the value of the security class parameter is lower than the target value and at least three entity data set for which the value of the security class parameter is at or above than the target value.

The security profile can include security practices and/or a security record of an entity. One or more input parameters indicative of the security profile of the entity can include: (a) an amount of capital investment in security of the entity; (b) a measure of employee training in security of the entity; (c) a measure of organization of a team dedicated to information security; and/or (d) an amount of budget dedicated to information security. One or more input parameters indicative of the security profile of the entity can include: (i) a number and/or severity of botnet infection instances of a computer system associated with the entity; (ii) a number of spam propagation instances originating from a computer network associated with the entity; (iii) a number of malware servers associated with the entity; (iv) a number of potentially exploited devices associated with the entity; (v) a number of hosts authorized to send emails on behalf of each domain associated with the entity; (vi) a determination of whether a DomainKeys Identified Mail (DKIM) record exists for each domain associated with the entity and/or a key length of a public key associated with a Domain Name System (DNS) record of each domain associated with the entity; (vii) an evaluation of a Secure Sockets Layer (SSL) certificate and/or a Transport Layer Security (TLS) certificate associated with a computer system of the entity; (viii) a number and/or type of service of open ports of a computer network associated with the entity; (ix) an evaluation of security-related fields of an header section of HTTP response messages of hosts associated with the entity; (x) a rate at which vulnerabilities are patched in a computer network associated with the entity; (xi) an evaluation of file sharing traffic originating from a computer network associated with the entity; and/or (xii) a number of lost records and/or sensitivity of information in the lost records in a data breach of a computer system associated with the entity.

The security class can be a security rating of the entity. The value of the security class parameter can be indicative of a security class above or below a target security rating. The statistical classifier can be: (i) a K-nearest neighbor algorithm, (ii) a support vector machine (SVM) model, or (iii) random forest classifier. Each entity data set can include two or more input parameters indicative of the security profile of the entity. The method can include, for a first input parameter of the two or more input parameters: determining a relationship between at least one value of the first input parameter and at least one value of a second input parameter; and storing the relationship in a database. The method can include determining relationships between a plurality of values of the first input parameter and a plurality of values of the second input parameter. The plurality of values of the first input parameter can include one or more values of the first input parameter and the plurality of values of the second input parameter can include one or more values of the second input parameter.

The method can include receiving values of the two or more input parameters for the particular entity; adjusting the value of the first input parameter of the two or more input parameters; determining the value of the second input parameter of the two or more input parameters based on the stored relationship in the database; using the trained statistical classifier on the adjusted value of the first input parameter and the determined value of the second input parameter to infer a value of the security class parameter indicative of the security class of the particular entity; comparing the value of the security class parameter to a target value to determine whether the adjustment of the value of the first input parameter results in a value of the security class parameter at, above, or below the target value. If the adjustment of the value of the first input parameter results in a value above the target value, the method can include generating a security improvement plan based on the adjusted value of the first input parameter and the determined value of the second input parameter, such that, if executed by the particular entity, increases the value of the security class parameter of the particular entity to or above the target value.

The security improvement plan can include a target value for at least one input parameter for the particular entity, in which the target value is different than the value of the at least one input parameter. The method can include presenting the security improvement plan via a user interface. The security improvement plan can include a prescription to adjust at least one of the input parameters. The method can include determining an explanation for the prescription using one or more explanation techniques selected from the group consisting of: (i) local interpretable model-agnostic explanation (LIME), (ii) high-precision model-agnostic explanation, (iii) Skater model interpretation, or (iv) random forest feature tweaking. The method can include presenting the explanation via the user interface. The method can include, if the adjustment of the value of the first input parameter results in the value of the security class parameter being at or above the target value, determining a target value for the first input parameter by: receiving two or more values of the first input parameter from two or more entity data sets of entities having a value of the security class parameter greater than the target value; determining a mean of the two or more values; generating a security improvement plan prescribing the mean value for the first input parameter of the particular entity.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1A is a flowchart of an exemplary method for statistical modeling of entities of a particular type. FIG. 1B is a diagram of the workflow for training the statistical model in FIG. 1A.

FIG. 2A is a flowchart of an exemplary method for using the trained statistical model. FIG. 2B is a flowchart of exemplary method for generating a security improvement plan. FIG. 2C is a diagram of the workflow for generating the security improvement plan of FIG. 2B.

FIG. 3 is a diagram of an exemplary security improvement plan for a particular entity.

FIG. 4 is a diagram of an exemplary computer system that may be used in implementing the systems and methods described herein.

DETAILED DESCRIPTION

Disclosed herein are exemplary embodiments of systems and methods for generating security improvement plans for entities. The security improvement plans can include values for one or more modifiable parameters (also referred to as “input parameters” herein) that a given entity can act upon to improve its security rating. Examples of security ratings and the determination of security ratings for entities can be found in commonly owned U.S. Pat. No. 9,973,524 issued May 15, 2018 and titled “Information Technology Security Assessment System,” the entirety of which is incorporated by reference herein. One example of a security rating (provided by BitSight Technologies, Inc., Boston, Mass.) has a scale from 300 (lowest) to 900 (highest). In some embodiments, lower security ratings signify high incidence of past security events and/or high risk of future security risk. Conversely, higher security ratings can signify low incidence of past security events and/or low risk of future security risk.

In the below disclosure, the following non-limiting example entity is utilized for illustrating the exemplary systems and methods described herein:

• • Corporation A is a financial services company (static parameter “industry”) having approximately 190 employees (static parameter “size”) in the northwest region of the United States (static parameter “location”).
  • Corporation A is seeking to increase its security rating from its current security rating of 560 to the target security rating of at least 720.

Similar Entities & Related Datasets

In some embodiments, the systems and methods described herein generate achievable security improvement plans by utilizing data associated with similar entities to the particular entity for which the plan is generated. In some embodiments, these similar entities can include those entities that have one or more static parameters that are shared with the particular entity. In some embodiments, the similarity between a first entity and a second entity is determined by a percentage of static parameters. For example, the first entity may share at least 50% of the static parameters with the second entity and therefore be considered to be “similar” to the second entity. In other examples, the first entity shares at least 70%, at least 80%, or at least 90% of the static parameters with the second entity to be considered “similar” to the second entity.

The below table provides an example set of entities sharing one or more static parameters with the exemplary “Corporation A,” as described above.

TABLE 1
List of entities, their associated static parameters, and similarity to
Corporation A. Note, in this example, that the similarity of the size
of entities may be determined by a predetermined category of sizes
for financial service companies (e.g., 1-50 employees, 51-200
employees, 201-500 employees, 501+ employees).
	Static Parameters {industry, size,
Entity	location}	Similarity
Corporation B	{financial services, 200 employees,	2/3
	southwest US}
Corporation F	{financial services, 150 employees,	3/3
	northwest US}
Corporation K	{financial services, 540 employees,	2/3
	northwest US}
Corporation Q	{financial services, 175 employees,	2/3
	southeast US}
Corporation S	{financial services, 50 employees,	1/3
	northeast US}

In the above example utilizing a 50% threshold in determining similarity between entities, only Corporation S lacks in similarity to Corporation A in most of the static parameters. Thus, only Corporation B, F, K, and Q would be used to train the statistical model to generate a security improvement panel for Corporation A, as described below.

In some embodiments, similar entities having security ratings above and below the target security rating are selected for generating the security improvement plan. Selecting entities having security ratings above the target security rating can contribute to determining which values for modifiable parameters that is likely to increase the particular entity's security rating. Selecting entities having security ratings below the target security rating can contribute to determining which values will not lead the particular entity to its target security rating (or above the target security rating). In some embodiments, similar entities having security ratings at, above, and below the target security rating are selected so that both of the above-described benefits are included in the generated security improvement plan. In some embodiments, a minimum number (e.g., at least three, at least 5, at least 10, etc.) of similar entities are selected for training the statistical model. In some embodiments, the number of similar entities having security ratings above the target security rating is approximately equal to the number of similar entities having security ratings below the target security rating. For example, approximately twenty similar entities may be selected having security ratings above the target security rating and approximately twenty similar entities may be selected below the security rating.

In some embodiments, for each of the similar entities, an entity data set is obtained. Continuing the above example, Table 2 lists the security ratings for the above ‘similar’ entities to Corporation A at a particular time.

TABLE 2
List of similar entities and their corresponding security ratings
at a particular time (e.g., 3 months ago, present date, etc.).
Entity        Security Rating
Corporation B 710
Corporation F 535
Corporation K 745
Corporation Q 680

In some embodiments, the security ratings of the similar entities include security ratings over a period, e.g., from a first time (e.g., 3 years ago, 1 year ago, 6 months ago, etc.) to a second time (e.g., 1 year ago, 6 months ago, present date, etc.). In some embodiments, the security ratings of entities may be averaged over some time period (e.g., within the last three months, last six months, last one year, etc.) to determine whether the entity should be selected. In some embodiments, other data related to security ratings can be obtained. For example, the other data can include data related to security events, components of the security ratings, analytics associated with the security ratings, etc. Examples of data related to security ratings can be found in U.S. patent application Ser. No. 16/360,641 titled “Systems and methods for forecasting cybersecurity ratings based on event-rate scenarios,”.

Training the Statistical Model

In some exemplary methods discussed herein, to determine values for the modifiable input parameters for the security improvement plan of a particular entity, a statistical model can be trained. In some embodiments, the statistical model can be trained on a plurality of entity data sets of entities similar to the particular entity. Further, the plurality of entity data sets can be selected such that the similar entities have security ratings both above and below the target security rating.

FIG. 1A is a flowchart illustrating a method 100 for statistical modeling of entities of a particular type. FIG. 1B is a diagram illustrating workflow 106 of training the statistical classifier 108. Step 102 of method 100 includes obtaining entity data including a plurality of entity datasets. Each entity data set 110 can be associated with a respective entity and include value(s) for one or more static parameters 112 indicative of a type of the entity. For example, the static parameters 112 can include entity size, entity industry, and/or entity location. The values of the static parameters 112 for each entity data set can indicate whether the type of the entity matches the particular type associated with the particular entity (see discussion above under heading “Similar Entities & Related Datasets”).

In some embodiments, each entity data set can include (i) values for one or more input parameters 114 indicating the security profile of the entity and/or (ii) a value of a security class parameter 116 indicating the security class of the entity based on the value(s) of the input parameter(s) 114. The security profile may include the security practices and/or security record of an entity. In some embodiments, the input parameters 114 can include one or more of:

• • an amount of capital investment in security of the entity;
  • a measure of employee training in security of the entity;
  • a measure of organization of a team dedicated to information security;
  • an amount of budget dedicated to information security;
  • a number and/or severity of botnet infection instances of a computer system associated with the entity;
  • a number of spam propagation instances originating from a computer network associated with the entity;
  • a number of malware servers associated with the entity;
  • a number of potentially exploited devices associated with the entity;
  • a number of hosts authorized to send emails on behalf of each domain associated with the entity;
  • a determination of whether a DomainKeys Identified Mail (DKIM) record exists for each domain associated with the entity and/or a key length of a public key associated with a Domain Name System (DNS) record of each domain associated with the entity;
  • an evaluation of a Secure Sockets Layer (SSL) certificate and/or a Transport Layer Security (TLS) certificate associated with a computer system of the entity;
  • a number and/or type of service of open ports of a computer network associated with the entity;
  • an evaluation of security-related fields of an header section of HTTP response messages of hosts associated with the entity;
  • a rate at which vulnerabilities are patched in a computer network associated with the entity;
  • an evaluation of file sharing traffic originating from a computer network associated with the entity; or
  • a number of lost records and/or sensitivity of information in the lost records in a data breach of a computer system associated with the entity.

In some embodiments, an entity data set can include two or more input parameters 114 (e.g., of those listed above). Thus, in some cases, the exemplary methods described herein can further include determining a relationship between a value of the first input parameter and a value of the second input parameter. This relationship can be stored in a database. For example, the number of botnet infections of an entity may be correlated with the number of potentially exploited devices associated with the entity. This correlation can be stored and referenced in the future. In some embodiments, the database includes the relationship between a plurality of values for the first input parameter and a plurality of values for the second input parameter. Relationships between values of the first and second parameters can be of a linear, non-linear, inverse, or other type. In some cases, the relationships can be stochastic.

In some embodiments, the security class parameter 116 of an entity is associated with, related to, or equal to the security rating of that entity (e.g., on a scale from 300 to 900, as provided by BitSight Technologies, Inc., Boston, Mass. and discussed above). For example, a first value of the security class parameter 116 is associated with, related to, or equal to a first security rating (e.g., 600); a second value of the security class parameter 116 is associated with, related to, or equal to a second security rating (e.g., 601); and so on. In some embodiments, the security class parameter 116 is associated with ranges of the security rating of the entity. For example, a first value of the security class parameter 116 is associated with, related to, or equal to a first security rating range (e.g., 600-649); a second value of the security class parameter 116 is associated with, related to, or equal to a second security rating (e.g., 650-659); and so on. In some embodiments, the value of the security class parameter 116 can indicate whether the security rating of the entity is at, above, or below a target security rating. For example, a first value of security class parameter is associated with, related to, or equal to a first set of security ratings at or above the target security rating (e.g., for a target security rating of 720, the first set of security ratings is 720-900); a second value of security class parameter is associated with, related to, or equal to a second set of security ratings below the target security rating (e.g., for a target security rating of 720, the second set of security ratings is 300-719).

In some embodiments, the method 100 can include selecting a target value for security class parameter 116 indicative of the security class for the particular entity. Having selected a target value, the plurality of entity data sets are chosen such that they include entity data set(s) for which the value of the security class parameter 116 is lower than the target value and entity data set(s) for which the value of the security class parameter 116 is greater than the target value. For example, if the target value of the security class parameter 116 (e.g., the security rating) for the particular entity is 720, then the data sets of one or more entities having a security rating less than 720 and the data sets of one or more entities having a security rating greater than 720 are selected for training the statistical classifier 108. In some cases, it can be beneficial to include entity data sets of having security class parameter values both above and below the target value in the training of the statistical model so that the generated security improvement plan, as discussed further below, includes values for one or more input parameters that can help the particular entity achieve the target security rating (or above the target security rating). Additionally or alternatively, the generated security improvement plan can provide values that can harm the particular entity's security rating (in other words, to illustrate for the particular entity ‘what not to do’ in their security practices).

Step 104 of method 100 includes training a statistical classifier to infer a value of the security class parameter indicative of the security class for the particular entity based on values of one or more of the input parameters indicative of a security profile of the particular entity. The training can include fitting the statistical classifier 108 to the plurality of entity data sets 110. Examples of the statistical classifier 108 can include any suitable statistical model for this use and can include any one of the following algorithms or models: a K-nearest neighbor algorithm; a support vector machine (SVM) model; or a decision tree-based model. For example, the decision tree-based model can be a random decision forest classifier (also known as a ‘random forest’). In some embodiments, the SVM model can include a radial basis function (RBF) kernel.

Generating Improvement Plans

FIG. 2A illustrates a method 200 for using the trained statistical classifier of method 100. In step 202 of method 200, values of input parameter(s) for the particular entity are received. In step 204 of method 200, the value of the first input parameter is adjusted (e.g., increased or decreased). In step 206, the value of the second input parameter is determined based on the stored relationship in the database. Referring to the example provided above, if there is an increased number of botnet infections (the value of the first parameter), then there is an expected increase in the number of potentially exploited devices (the value of the second parameter) based on the stored relationship. Therefore, the number of potentially exploited devices is determined to increase as well.

In step 206, the trained statistical classifier (see discussion above under heading “Training the Statistical Model”) can be used on the adjusted value of the first input parameter and the determined value of the second input parameter to infer a value of the security class parameter of the entity. In step 208, the value of the security class parameter can be compared to a target value to determine whether the adjustment of the value of the first input parameter results in a value of the security class parameter above or below the target value. For example, the classifier may infer a value of the security class parameter (e.g., the security rating) to be 685 based on an increased number of botnet infections. If the target value of the security class parameter (e.g., the security rating) is 720, then the adjustment results in a value of the security class parameter below the target value (e.g., 685 compared to 720).

FIG. 2B illustrates a method 201 for generating a security improvement plan for the particular entity. FIG. 2C illustrates a workflow 211 for generating the security improvement plan using the trained classifier (refer to exemplary method 200 for description related to steps 202 through 210). In step 212 of method 201, a security improvement plan 222 can be generated for the particular entity. The security improvement plan can be based on the adjusted value of the first input parameter and the determined value of the second input parameter. If executed by the particular entity, the security improvement plan 222 can increase the value of the security class parameter of the particular entity to or above the target value. In an ideal scenario, the particular entity is expected to execute the generated security improvement plan 222 by attempting to attain each of the values of the modifiable input parameters. In some embodiments, the security improvement plan 222 is presented to a user (e.g., company representative, insurance representative, etc.) via a user interface. FIG. 3 is a diagram of an exemplary security improvement plan 214 for the particular entity.

In some embodiments, exemplary method 200 and/or exemplary method 201 can include determining the mean of two or more values of the first input parameter from two or more entity data sets of entities having a value of the security class parameter greater than the target value. Methods 200 and/or 201 can include the generation of a security improvement plan 222 prescribing the mean value for the first input parameter of the particular entity. In some embodiments, this technique can be repeated for each input parameter that is found to contribute to an improved security rating for the entity. For example, the contribution of an input parameter to the security rating can be determine by steps 202-210 of methods 200 or 201.

In some embodiments, the security improvement plan for the particular entity can include a prescription to adjust at least one input parameter. It can be beneficial to provide explanations to the particular entity as to why modifying the values of which parameters helps the entity achieve the desired security rating. The method 201 can include one or more explanation techniques, e.g., local interpretable model-agnostic explanation (LIME), high-precision model-agnostic explanation (referred to as ‘anchors’), Skater model interpretation, random forest feature tweaking, etc. In some embodiments, the explanations can be presented to the user via the user interface.

Computer-Based Implementations

In some examples, some or all of the processing described above can be carried out on a personal computing device, on one or more centralized computing devices, or via cloud-based processing by one or more servers. In some examples, some types of processing occur on one device and other types of processing occur on another device. In some examples, some or all of the data described above can be stored on a personal computing device, in data storage hosted on one or more centralized computing devices, or via cloud-based storage. In some examples, some data are stored in one location and other data are stored in another location. In some examples, quantum computing can be used. In some examples, functional programming languages can be used. In some examples, electrical memory, such as flash-based memory, can be used.

FIG. 4 is a block diagram of an example computer system 400 that may be used in implementing the technology described in this document. General-purpose computers, network appliances, mobile devices, or other electronic systems may also include at least portions of the system 400. The system 400 includes a processor 410, a memory 420, a storage device 430, and an input/output device 440. Each of the components 410, 420, 430, and 440 may be interconnected, for example, using a system bus 450. The processor 410 is capable of processing instructions for execution within the system 400. In some implementations, the processor 410 is a single-threaded processor. In some implementations, the processor 410 is a multi-threaded processor. The processor 410 is capable of processing instructions stored in the memory 420 or on the storage device 430.

The memory 420 stores information within the system 400. In some implementations, the memory 420 is a non-transitory computer-readable medium. In some implementations, the memory 420 is a volatile memory unit. In some implementations, the memory 420 is a nonvolatile memory unit.

The storage device 430 is capable of providing mass storage for the system 400. In some implementations, the storage device 430 is a non-transitory computer-readable medium. In various different implementations, the storage device 430 may include, for example, a hard disk device, an optical disk device, a solid-date drive, a flash drive, or some other large capacity storage device. For example, the storage device may store long-term data (e.g., database data, file system data, etc.). The input/output device 440 provides input/output operations for the system 400. In some implementations, the input/output device 440 may include one or more of a network interface devices, e.g., an Ethernet card, a serial communication device, e.g., an RS-232 port, and/or a wireless interface device, e.g., an 802.11 card, a 3G wireless modem, or a 4G wireless modem. In some implementations, the input/output device may include driver devices configured to receive input data and send output data to other input/output devices, e.g., keyboard, printer and display devices 460. In some examples, mobile computing devices, mobile communication devices, and other devices may be used.

In some implementations, at least a portion of the approaches described above may be realized by instructions that upon execution cause one or more processing devices to carry out the processes and functions described above. Such instructions may include, for example, interpreted instructions such as script instructions, or executable code, or other instructions stored in a non-transitory computer readable medium. The storage device 430 may be implemented in a distributed way over a network, such as a server farm or a set of widely distributed servers, or may be implemented in a single computing device.

Although an example processing system has been described in FIG. 4, embodiments of the subject matter, functional operations and processes described in this specification can be implemented in other types of digital electronic circuitry, in tangibly-embodied computer software or firmware, in computer hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible nonvolatile program carrier for execution by, or to control the operation of, data processing apparatus. Alternatively or in addition, the program instructions can be encoded on an artificially generated propagated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal that is generated to encode information for transmission to suitable receiver apparatus for execution by a data processing apparatus. The computer storage medium can be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of one or more of them.

The term “system” may encompass all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. A processing system may include special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit). A processing system may include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them.

A computer program (which may also be referred to or described as a program, software, a software application, a module, a software module, a script, or code) can be written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages, and it can be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

The processes and logic flows described in this specification can be performed by one or more programmable computers executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).

Computers suitable for the execution of a computer program can include, by way of example, general or special purpose microprocessors or both, or any other kind of central processing unit. Generally, a central processing unit will receive instructions and data from a read-only memory or a random access memory or both. A computer generally includes a central processing unit for performing or executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Moreover, a computer can be embedded in another device, e.g., a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a Global Positioning System (GPS) receiver, or a portable storage device (e.g., a universal serial bus (USB) flash drive), to name just a few.

Computer readable media suitable for storing computer program instructions and data include all forms of nonvolatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

To provide for interaction with a user, embodiments of the subject matter described in this specification can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input. In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user's user device in response to requests received from the web browser.

Embodiments of the subject matter described in this specification can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the subject matter described in this specification, or any combination of one or more such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (“LAN”) and a wide area network (“WAN”), e.g., the Internet.

The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

While this specification contains many specific implementation details, these should not be construed as limitations on the scope of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a sub-combination or variation of a sub-combination.

Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

Particular embodiments of the subject matter have been described. Other embodiments are within the scope of the following claims. For example, the actions recited in the claims can be performed in a different order and still achieve desirable results. As one example, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In certain implementations, multitasking and parallel processing may be advantageous. Other steps or stages may be provided, or steps or stages may be eliminated, from the described processes. Accordingly, other implementations are within the scope of the following claims.

Terminology

The phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting.

The term “approximately”, the phrase “approximately equal to”, and other similar phrases, as used in the specification and the claims (e.g., “X has a value of approximately Y” or “X is approximately equal to Y”), should be understood to mean that one value (X) is within a predetermined range of another value (Y). The predetermined range may be plus or minus 20%, 10%, 5%, 3%, 1%, 0.1%, or less than 0.1%, unless otherwise indicated.

The indefinite articles “a” and “an,” as used in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.” The phrase “and/or,” as used in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and/or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and/or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

As used in the specification and in the claims, “or” should be understood to have the same meaning as “and/or” as defined above. For example, when separating items in a list, “or” or “and/or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of” or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used shall only be interpreted as indicating exclusive alternatives (i.e. “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

As used in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and/or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

The use of “including,” “comprising,” “having,” “containing,” “involving,” and variations thereof, is meant to encompass the items listed thereafter and additional items.

Use of ordinal terms such as “first,” “second,” “third,” etc., in the claims to modify a claim element does not by itself connote any priority, precedence, or order of one claim element over another or the temporal order in which acts of a method are performed. Ordinal terms are used merely as labels to distinguish one claim element having a certain name from another element having a same name (but for use of the ordinal term), to distinguish the claim elements.

Claims

1. A computer-implemented method for executing a security improvement plan for a particular organization, the method comprising: receiving an organization data set for the particular organization, the organization data set comprising: (i) a value for at least one static parameter indicative of a type of the organization;(ii) a value for at least two input parameters indicative of a security profile of the particular organization; and(iii) a value of a security class parameter indicative of a security class of the particular organization;adjusting a value of a first input parameter of the input parameters;determining a value of a second input parameter of the input parameters based on a relationship between the value of the first input parameter and the value of the second input parameter;using a trained statistical classifier on the adjusted value of the first input parameter and the determined value of the second input parameter to infer an adjusted value of the security class parameter; andif the adjustment of the value of the first input parameter results in an increased value of the security class parameter, selecting the first input parameter for modification in a security improvement plan based on the adjusted value of the first input parameter, wherein modification of the first input parameter by the particular organization implies an increase in the value of the security class parameter of the particular organization.

2. The method of claim 1, further comprising: comparing the adjusted value of the security class parameter to a target value to determine whether the adjustment of the value of the first input parameter results in a value of the security class parameter at, above, or below the target value.

3. The method of claim 2, wherein selecting the first input parameter for modification comprises: executing the security improvement plan for the particular organization, wherein execution of the security improvement plan results in an increase in the value of the security class parameter of the particular organization at or above the target value, wherein generation of the security improvement plan is based on the adjusted value of the first input parameter.

4. The method of claim 1, wherein the security improvement plan comprises a target value for at least one input parameter for the particular organization, the target value being different than the value of the at least one input parameter.

5. The method of claim 1, further comprising: presenting the security improvement plan via a user interface.

6. The method of claim 1, wherein the security improvement plan includes a prescription to adjust at least one of the input parameters, the method further comprising: determining an explanation for the prescription using one or more explanation techniques selected from the group consisting of: (i) local interpretable model-agnostic explanation (LIME), (ii) high-precision model-agnostic explanation, (iii) Skater model interpretation, or (iv) random forest feature tweaking.

7. The method of claim 6, further comprising: presenting the explanation via the user interface.

8. The method of claim 1, further comprising: determining a target value for the first input parameter by: receiving two or more values of the first input parameter from two or more organization data sets of entities having a value of the security class parameter greater than the target value; anddetermining a mean of the two or more values,wherein selecting the first input parameter for modification comprises using the mean value for the first input parameter of the particular organization.

9. The method of claim 1, wherein the relationship between at least one value of the first input parameter and at least one value of a second input parameter is stored in a database, and wherein the method comprises: retrieving the stored relationship from the database.

10. The method of claim 1, wherein the static parameters comprise at least one of (i) organization size, (ii) organization industry, and/or (iii) organization location.

11. The method of claim 1, wherein the security profile comprises security practices and/or a security record of an organization.

12. The method of claim 1, wherein the one or more input parameters indicative of the security profile of the organization comprise at least one of: an amount of capital investment in security of the organization;a measure of employee training in security of the organization;a measure of organization of a team dedicated to information security; oran amount of budget dedicated to information security.

13. The method of claim 1, wherein the one or more input parameters indicative of the security profile of the organization comprise at least one of: a number and/or severity of botnet infection instances of a computer system associated with the organization;a number of spam propagation instances originating from a computer network associated with the organization;a number of malware servers associated with the organization;a number of potentially exploited devices associated with the organization;a number of hosts authorized to send emails on behalf of each domain associated with the organization;a determination of whether a DomainKeys Identified Mail (DKIM) record exists for each domain associated with the organization and/or a key length of a public key associated with a Domain Name System (DNS) record of each domain associated with the organization;an evaluation of a Secure Sockets Layer (SSL) certificate and/or a Transport Layer Security (TLS) certificate associated with a computer system of the organization;a number and/or type of service of open ports of a computer network associated with the organization;an evaluation of security-related fields of an header section of HTTP response messages of hosts associated with the organization;a rate at which vulnerabilities are patched in a computer network associated with the organization;an evaluation of file sharing traffic originating from a computer network associated with the organization; ora number of lost records and/or sensitivity of information in the lost records in a data breach of a computer system associated with the organization.

14. The method of claim 1, wherein the statistical classifier was trained by: obtaining organization data including a plurality of organization data sets, each organization data set associated with a respective organization and including: (i) a value for at least one static parameter indicative of a type of the organization, wherein the values of the static parameter indicates that the type of the organization matches the particular type;(ii) a value for at least one input parameter indicative of a security profile of the organization;(iii) a value of a security class parameter indicative of a security class of the organization based on the value of the at least one input parameter;training the statistical classifier to infer a value of the security class parameter indicative of the security class of a particular organization of the particular type based on values of the at least one input parameter indicative of a security profile of the particular organization, wherein training the statistical classifier comprises fitting the statistical classifier to the plurality of organization data sets.

15. The method of claim 14, wherein each organization data set includes two or more input parameters indicative of the security profile of the organization, the method further comprising: for a first input parameter of the two or more input parameters, determining the relationship between at least one value of the first input parameter and at least one value of a second input parameter.

16. The method of claim 15, further comprising: determining relationships between a plurality of values of the first input parameter and a plurality of values of the second input parameter, wherein the plurality of values of the first input parameter comprises the at least one value of the first input parameter and the plurality of values of the second input parameter comprises the at least one value of the second input parameter.

17. The method of claim 14, wherein the values of two or more static parameters for each of the organization data sets indicate that the type of the organization matches the particular type.

18. The method of claim 14, further comprising: selecting a target value for the security class parameter indicative of the security class for the particular organization,wherein the plurality of organization data sets includes at least one organization data set for which the value of the security class parameter is lower than the target value and at least one organization data set for which the value of the security class parameter is at or above than the target value.

19. The method of claim 18, wherein the plurality of organization data sets includes at least three organization data sets for which the value of the security class parameter is lower than the target value and at least three organization data sets for which the value of the security class parameter is at or above than the target value.

20. The method of claim 1, wherein the security class is a security rating of the particular organization.