International Business Machines Corporation

A9.com, Inc.

Shay Artzi

Marco Pistoia

Julian Timothy Dolby

Frank Tip

Manu Sridharan

Omer Tripp

Adrian Boteanu

Adam Kiezun

Ryan Berg

Robert Wiener

Raju Matta

Takaaki Tateishi

John Peyton

Emily Dutile

Salvatore Angelo Guarnieri

John Thomas Peyton, Jr

Gaurav Gupta

Michael James Morris

Joshua Christian Snyder

Simon H Jensen

Systems and methods are disclosed for a system to provide an interface that is dynamic and that provides selectable links in response to a query for products in an electronic marketplace, where the selectable links are titled with the query and portions of reviews for products associated with the query. The system is configured to select feedback for items purchased from an electronic marketplace. Descriptors from the feedback are generated. In response to a query for the one or more of the items in the electronic marketplace, a determination is made that portions of the descriptors provide detail responsive to the query. An interface is displayed including selectable links titled with the query in combination with the portions of the descriptors. In response to selection of one of the selectable links, a portion of the items are displayed.

Indexing and presenting content using latent interests

Various embodiments of systems and methods for mapping review articles to locations in a marketplace catalog. Various embodiments include training a system using product reviews and images from the marketplace as training data. Product reviews and images can be provided to autoencoders to generate respective short codes. The product review short codes can be combined with the product image short codes to generate review short codes. These review short codes can be provided as training data for an attribute assignment network while associated attributes (e.g., attributes associated with related products) can be target outputs for the attribute assignment network. After the system is sufficiently trained, a review article can be obtained from the publishing platform. This review article can be provided to the trained system to receive predicted attribute assignments. These predicted attribute assignments can be effective to map the review article to locations in a marketplace catalog.

Mapping content to an item repository

Systems and methods are disclosed for refining the accuracy of network searches by supplementing existing keywords and key phrases in an e-commerce catalog or other database with aggregated and analyzed additional, external data. The internet or another network can be crawled for identifiers which point to entries in the catalog or other database, and, subject to third-party use restrictions, data and metadata can be extracted to enrich the existing keywords and key phrases. The extracted external content may be processed by machine learning techniques in order to find similar entries in the original catalog or database. Categorizing and indexing the entries further improves search recall, including clustering via processing word embeddings.

Refined search query results through external content aggregation and application

A novel system, computer program product, and method are disclosed for feedback-directed automated test generation for programs, such as JavaScript, in which execution is monitored to collect information that directs the test generator towards inputs that yield increased coverage. Several instantiations of the framework are implemented, corresponding to variations on feedback-directed random testing, in a tool called Artemis.

Automated testing of applications with scripting code

A method includes determining selected global variables in a program for which flow of the selected global variables through the program is to be tracked. The selected global variables are less than all the global variables in the program. The method includes using a static analysis performed on the program, tracking flow through the program for the selected global variables. In response to one or more of the selected global variables being used in security-sensitive operations in the flow, use is analyzed of each one of the selected global variables in a corresponding security-sensitive operation. In response to a determination the use may be a potential security violation, the potential security violation is reported. Apparatus and computer program products are also disclosed.

Global variable security analysis

Disclosed is a novel computer implemented system, on demand service, computer program product and a method that leverages combined concrete and symbolic execution and several fault-localization techniques to automatically detects failures and localizes faults in PHP Hypertext Preprocessor ('PHP') Web applications.

Fault localization using condition modeling and return value modeling

The present invention provides a system, computer program product and a computer implemented method for prioritizing code fragments based on the use of a software oracle and on a correlation between the executed code fragments and the output they produce. Also described is a computer-implemented method generates additional user inputs based on execution information associated with path constraints and based on information from the oracle. Advantageously, the embodiment is useful in a test generation tool that generated many similar inputs when a failure-inducing input is found, in order to enhance fault localization. Further, described is a computer-implemented flow for extending the existing idea of concolic testing to applications that interact with persistent state.

Fault detection and localization in dynamic software applications

Systems and methods are provided for creating a data structure associated with a software application that is based on at least one framework. According to the method, at least one Java Server Page file associated with the software application is analyzed. The Java Server Page (JSP) file includes at least one call to at least one library tag, and at least one Expression Language (EL) expression. A set of tag library usage information for the JSP file is generated based. The set of tag library usage information includes at least one variable, and a value of the at least one variable created by the at least one call. The EL expression is evaluated based on the variable and the value of the variable. A data structure is created for a static analysis engine based on EL expression. The data structure includes at least one Java expression representing the EL expression.

Generating specifications for expression language expressions and tag libraries

Fault detection and localization in dynamic software applications requiring user inputs and persistent states

The present invention provides a system, computer program product, and a computer implemented method for analyzing a set of two or more communicating applications. The method includes executing a first application, such as a client application, and executing a second application, such as a server application. The applications are communicating with each other. A correlation is recorded between the applications and an execution characteristic exhibited on execution. An oracle is used to determine an analysis of the first application that has been executed. The execution of the first application causes a change of state in the second application and/or a change control flow in the second application. Code fragment in the first application and/or the second application are prioritized based on an evaluation produced by the oracle, and based on the correlation between the code fragments that have been executed and the execution characteristic exhibited by the code fragments.

Detecting and localizing security vulnerabilities in client-server application

The present invention provides a system, computer program product, and a computer implemented method for analyzing a set of two or more communicating applications. The method begins with receiving a first second application that communicates with each other during execution. Next, an initial input for executing the first application and the second application is received. The initial input is added to a set of inputs. An iterative execution loop is performed at least once. The loop begins with selecting inputs out of the set of inputs for execution. Next, using the selected inputs, the first and/or the second application is executed while information regarding the execution and information communicated to the other application are recorded. A set of one or more new application inputs for either applications is generated based the second application recorded information and the first application information. These new inputs are added to the set of inputs.

Generating inputs for client-server programs for fault-detection and localization

Systems and methods are provided for creating a data structure associated with a software application that is based on at least one framework. According to the method, source code and at least one configuration file of the software application is analyzed by at least one framework-specific processor so as to determine entry point information indicating entry points in the source code, request attribute access information indicating where attributes attached to a request data structure are read and written, and forward information indicating forwards performed by the software application. A data structure for a static analysis engine is created based on this information. The data structure includes a list of synthetic methods that model framework-related behavior of the software application, and a list of entry points indicating the synthetic methods and/or application methods of the software application that can be invoked by the framework.

Generating specifications of client-server applications for static analysis

Disclosed is a novel computer implemented system, on demand service, computer program product and a method for fault-localization techniques that apply statistical analyses to execution data gathered from multiple tests. The present invention determines the fault-localization effectiveness of test suites generated according to several test-generation techniques based on combined concrete and symbolic (concolic) execution. These techniques are evaluated by applying the Ochiai fault-localization technique to generated test suites in order to localize 35 faults in four PHPWeb applications. The results show that the test-generation techniques under consideration produce test suites with similar high fault-localization effectiveness, when given a large time budget.

Growing community of inventors

Brookline, MA, United States of America

Shay Artzi