Abstract: An e-commerce platform may be subject to fraudulent activity. In some embodiments, a graph is used to represent the connection between users of the e-commerce platform. The graph may be built and maintained using a belief propagation algorithm. The graph may be used to assign, to each user, the probability that the user is fraudulent. In some embodiments, the use of the graph and updating the graph are two independent operations: one performed in real-time, and the other performed offline. In some embodiments, additional information may be requested from a user if the user is assigned a high probability of being fraudulent, in order to try to better determine whether or not the user is a fraudulent user. The embodiments are not limited to an e-commerce platform, but may generally apply to any computer system in which there is a desire to detect fraudulent activity.

Abstract: Computer-implemented methods and systems for colour-based image tagging and colour-based searching. The method may include identifying, using image analysis, one or more dominant colours of a product based on an image of the product and receiving selection of at least one of the one or more dominant colours. In response to receiving the selection of the at least one of the one or more dominant colours, a search for products matching the at least one of the one or more dominant colours may be initiated to obtain one or more results of the searching, the one or more results including at least one product matching the at least one of the one or more dominant colours.

Abstract: A data partitioning system receives an input dataset for e-commerce products, each sample containing attributes and associated values for each product including at least an image; represents each sample as a node on a graph to provide a graph of nodes for the dataset; measures a relative similarity distance between each pair of nodes based on comparing at least image values for the attributes; determines for each pair of nodes whether they are related if the similarity distance between them is below a defined threshold, and if related, generate an edge between them on the graph; group the connected nodes into a first or a second group such that the grouped nodes have no edges connecting them to nodes in the other group and have a shortest relative similarity distance with each other. The groups are used as training dataset and testing data sets for a supervised machine learning classifier.

Abstract: Computer-implemented methods and systems for colour-based image tagging and colour-based searching. The method may include identifying, using image analysis, one or more dominant colours of a product based on an image of the product and receiving selection of at least one of the one or more dominant colours. In response to receiving the selection of the at least one of the one or more dominant colours, a search for products matching the at least one of the one or more dominant colours may be initiated to obtain one or more results of the searching, the one or more results including at least one product matching the at least one of the one or more dominant colours.

Abstract: A data partitioning system receives an input dataset for e-commerce products, each sample containing attributes and associated values for each product including at least an image; represents each sample as a node on a graph to provide a graph of nodes for the dataset; measures a relative similarity distance between each pair of nodes based on comparing at least image values for the attributes; determines for each pair of nodes whether they are related if the similarity distance between them is below a defined threshold, and if related, generate an edge between them on the graph; group the connected nodes into a first or a second group such that the grouped nodes have no edges connecting them to nodes in the other group and have a shortest relative similarity distance with each other. The groups are used as training dataset and testing data sets for a supervised machine learning classifier.

Abstract: A method at a computing device for adaptive product categorization, the method including monitoring adoption of custom categories by merchants, the custom categories being custom additions to a standard taxonomy; clustering the custom categories to form clusters having associated cluster values; determining that a cluster value for a particular one of the custom categories exceeds a threshold; and adapting the standard taxonomy based on the particular one of the custom categories.

Abstract: Methods and systems of securely exchanging customer data are provided. The exchange may be done in privacy-preserving and/or privacy-protective fashion. Customer data from multiple merchants, including transaction data for a first product between a first customer and a first merchant, is maintained in a customer data pool by an e-commerce platform. When a second merchant associated with a second product requests new customer data, a product (i.e., the first product) relevant to the second product is identified. A collection of customers who indicated interest in the first product (including the first customer) are identified from the customer data pool, and a target audience is generated proportional to customer data exchanged by the second merchant with the customer data pool. Communications from the second merchant are then provided to the target audience without disclosing their customer data to the second merchant.

Abstract: An e-commerce platform may be subject to fraudulent activity. In some embodiments, a graph is used to represent the connection between users of the e-commerce platform. The graph may be built and maintained using a belief propagation algorithm. The graph may be used to assign, to each user, the probability that the user is fraudulent. In some embodiments, the use of the graph and updating the graph are two independent operations: one performed in real-time, and the other performed offline. In some embodiments, additional information may be requested from a user if the user is assigned a high probability of being fraudulent, in order to try to better determine whether or not the user is a fraudulent user. The embodiments are not limited to an e-commerce platform, but may generally apply to any computer system in which there is a desire to detect fraudulent activity.

Abstract: Systems and methods are provided for automatically making a decision as to whether or not to over an installment payment (IP) option to a customer at checkout at an online retailer for goods or services. The IP decision is made, at least in part, based on the specific product or products being purchased. In some cases, it is also based on merchant-specific information and/or customer specific information.

Abstract: Methods and systems for managing messaging campaigns are disclosed. A set of parameters are received for a proposed messaging campaign. Based on a trust indicator associated with the proposed messaging campaign, it is determined that the proposed messaging campaign should be blocked. At least one tracked message is transmitted to at least one selected recipient, where the tracked message is at least a subset of messages for the proposed messaging campaign. If a rejection metric associated with the tracked message passes a rejection threshold, messages are transmitted to the remaining recipients.

Abstract: Systems and methods of transmitting data over authorized transmission channels are provided. Data for transmission is segregated into subsets such that the subsets may be processed by multiple authorized transmission channels, thus providing flexibility in that data subsets may have differing characteristics that would allow transmission over different authorized transmission channels. The data transmission systems and methods may be implemented in an e-commerce environment to process transaction data via payment gateways.

Abstract: Systems and methods are provided for automatically making a decision as to whether or not to over an installment payment (IP) option to a customer at checkout at an online retailer for goods or services. The IP decision is made, at least in part, based on the specific product or products being purchased. In some cases, it is also based on merchant-specific information and/or customer specific information.