Abstract: Systems and methods are disclosed herein for classifying records, such as product records, using a machine learning algorithm. After training a classification model according to a machine learning algorithm using an initial training set, records are classified and high confidence classifications identified. Remaining classifications are submitted to a crowdsourcing forum that validates or invalidates the classifications or marks them as to unclear to evaluate. Invalidated classifications are automatically analyzed to identify one or both of classification values and categories having a high proportion of invalidated classifications. Requests are transmitted to analysts to generate training data that is added to the training set. The process of classifying records and obtaining crowdsourced validation thereof may then repeat. High confidence classifications may be identified using an accuracy model trained to relate an accuracy percentage to a confidence score output by the classification model.

Abstract: Systems and methods are disclosed herein for classifying records, such as product records, using a machine learning algorithm. After training a classification model according to a machine learning algorithm using an initial training set, records are classified and high confidence classifications identified. Remaining classifications are submitted to a crowdsourcing forum that validates or invalidates the classifications or marks them as to unclear to evaluate. Invalidated classifications are automatically analyzed to identify one or both of classification values and categories having a high proportion of invalidated classifications. Requests are transmitted to analysts to generate training data that is added to the training set. The process of classifying records and obtaining crowdsourced validation thereof may then repeat.

Abstract: The present disclosure extends to methods, systems, and computer program products for updating a merchant database with new items automatically or with minimal human involvement. In operation, methods and systems disclosed use a pipeline of classification models to quantify new product information and create an accurate classification for the new product item.

Abstract: The present disclosure extends to methods, systems, and computer program products for updating a database with new products by classifying the new products within a hierarchy, and then using the hierarchy to improve the classification by including other product types within the classification for the new products.

Abstract: The present disclosure extends to methods, systems, and computer program products for automatically determining key words within item information with product types, and classifying new items within product types within a merchant's database.

Abstract: The present disclosure extends to methods, systems, and computer program products for updating a merchant database with new product items and placing the new product items within a hierarchy of existing merchant product offerings. In operation, the new product is represented by a title and description that can be semantically classified using a plurality of classification models and reviewed by users for accuracy.

Abstract: The present disclosure extends to methods, systems, and computer program products for updating a merchant database with new products in an optimized manner using both computer based classification models and human involvement in a smart crowd source environment.

Abstract: Systems and methods are disclosed herein for performing classification of documents or performing other tasks based on rules. The rules may include context rules that define a mapping that relates a value and context in a document to an attribute to which the value corresponds. Products are selected for labeling with attributes by identifying patterns, e.g. values and contexts that are not covered by a current rule set. Those products having a highest score are selected for labeling in a crowd sourcing forum, where the score is based on the number of non-covered patterns and a frequency of occurrence of the non-covered patterns in a document corpus. Proposed rules are generated for frequently occurring patterns and submitted to analysts for one or both of completion and validation. Proposed rules may include a proposed attribute for a frequently occurring value and corresponding context.

Abstract: A crowd-sourcing system performs a method that includes generating a request that a user provide candidate media data to the crowd-sourcing system. The candidate media is descriptive of a product, and the request includes information about the product. The crowd-sourcing system transmits the request to a user system and receives the candidate media data from the user system. The crowd-sourcing system selects the candidate media data as stock media data for the product. The selection of the candidate media data may be based on a selection criterion, such as a grade of the candidate media data determined by the crowd-sourcing system. The crowd-sourcing system provides a reward to the user based on the selecting of the candidate media data as the stock media data for the product.

Abstract: A crowd-sourcing system performs a method that includes generating a request that a user provide candidate media data to the crowd-sourcing system. The candidate media is descriptive of a product, and the request includes information about the product. The crowd-sourcing system transmits the request to a user system and receives the candidate media data from the user system. The crowd-sourcing system selects the candidate media data as stock media data for the product. The selection of the candidate media data may be based on a selection criterion, such as a grade of the candidate media data determined by the crowd-sourcing system. The crowd-sourcing system provides a reward to the user based on the selecting of the candidate media data as the stock media data for the product.

Abstract: In a method of enabling a second system to manage selected anonymous users of a first system in a performance of a task, a rule is received from a second system, the rule specifying a criterion for a selection of a user from a plurality of users of the first system. A user is selected from the plurality of users, the selecting based on the criterion, the selecting including searching a record of a database, the record containing data pertaining to an activity of the user with respect to the first system, the database being inaccessible to the second system. A request is forwarded from the second system to the user, the request requesting that the user perform a task. A result is sent to the second system, the result including information pertaining to a performance of the task by the user.

Abstract: A crowd-sourcing system performs a method that includes generating a request that a user provide candidate media data to the crowd-sourcing system. The candidate media is descriptive of a product, and the request includes information about the product. The crowd-sourcing system transmits the request to a user system and receives the candidate media data from the user system. The crowd-sourcing system selects the candidate media data as stock media data for the product. The selection of the candidate media data may be based on a selection criterion, such as a grade of the candidate media data determined by the crowd-sourcing system. The crowd-sourcing system provides a reward to the user based on the selecting of the candidate media data as the stock media data for the product.