The Toronto-dominion Bank

Yahoo! Inc.

Microsoft Technology Licensing, LLC

Excalibur Ip, LLC

Tomi Johan Poutanen

Maksims Volkovs

Jean-Marc Langlois

Arkady Borkovsky

Douglas Wesley Cook

Hongyuan Zha

Paige Elyse Dickie

Kin Kwan Leung

Barum Rho

Jianjin Dong

Jonathan Forbes

The disclosed embodiments include computer-implemented apparatuses and processes that dynamically predict future occurrences of events using adaptively trained artificial-intelligence processes and contextual data. For example, an apparatus may generate an input dataset based on first interaction data and contextual data associated with a prior temporal interval, and may apply an adaptively trained, gradient-boosted, decision-tree process to the input dataset. Based on the application of the adaptively trained, gradient-boosted, decision-tree process to the input dataset, the apparatus may generate output data representative of a predicted likelihood of an occurrence of an event during a future temporal interval, which may be separated from the prior temporal interval by a corresponding buffer interval. The apparatus may also transmit a portion of the generated output data to a computing system, and the computing system may be configured to generate or modify second interaction data based on the portion of the output data.

Prediction of future occurrences of events using adaptively trained artificial-intelligence processes and contextual data

The disclosed embodiments include computer-implemented processes that flexibly and dynamically analyze a machine learning process, and that generate analytical output characterizing an operation of the machine learning process across multiple analytical periods. For example, an apparatus may receive an identifier of a dataset associated with the machine learning process and feature data that specifies an input feature of the machine learning process. The apparatus may access at least a portion of the dataset based on the received identifier, and obtain, from the accessed portion of the dataset, a feature vector associated with the machine learning process. The apparatus may generate a plurality of modified feature vectors based on the obtained feature vector, and based on an application of the machine learning process to the obtained and modified feature vectors, generate and transmit, to a device, first explainability data associated with the specified input feature for presentation within a digital interface.

Dynamic analysis and monitoring of machine learning processes

A document analysis system trains a document embedding model configured to receive a set of word embeddings for an ordered set of words in a document and generate a document embedding for the document. The document embedding is a representation of the document in a latent space that characterizes the document with respect to properties such as structure, content, and sentiment. The document embedding may represent a prediction of a set of words that follow the last word in the ordered set of words of the document. The document embedding model may be associated with a convolutional neural network (CNN) architecture that includes one or more convolutional layers. The CNN architecture of the document embedding model allows the document analysis system to overcome various difficulties of existing document embedding models, and allows the document analysis system to easily process variable-length documents that include a variable number of words.

Learning document embeddings with convolutional neural network architectures

One embodiment is a method executed by a computer system that applies collaborative filtering to provide a recommendation to a user. The method includes retrieving a binary matrix that includes rows and columns of binary data for preferences of users on items; applying a neighborhood-based approach to convert the binary matrix into a neighbor-score matrix; applying a factorization to approximate the neighbor-score matrix with a product of lower rank matrices; calculating a user factor and an item factor based on the factorization; calculating scores for user-item pairs by computing a dot product between the user factor and the item factor; sorting the scores of the user-item pairs to generate the recommendation to the user; and providing the recommendation to a general-purpose computer of the user.

Method and system for making recommendations from binary data using neighbor-score matrix and latent factors

A recommendation system generates recommendations for an online system using one or more neural network models that predict preferences of users for items in the online system. The neural network models generate a latent representation of a user and of a user that can be combined to determine the expected preference of the user to the item. By using neural network models, the recommendation system can generate predictions in real-time for new users and items without the need to re-calibrate the models. Moreover, the recommendation system can easily incorporate other forms of information other than preference information to generate improved preference predictions by including the additional information to generate the latent description of the user or item.

End-to-end deep collaborative filtering

In one embodiment, a data structure comprises: a primary index comprising one or more position-block references; and one or more position blocks sequentially following the primary index, wherein: each one of the position-block references corresponds to one of the position blocks; and each one of the position blocks comprises: a secondary index comprising one or more position-data references; and one or more sets of positions sequentially following the secondary index, wherein each one of the position-data references corresponds to of one of the sets of positions in the position block. In one embodiment, an instance of the data structure is stored in a computer-readable memory and accessible by an application executed by a process.

Encoding and accessing position data

A method and apparatus for using network traffic logs for search enhancement is disclosed. According to one embodiment, network usage is tracked by generating log files. These log files among other things indicate the frequency web pages are referenced and modified. These log files or information from these log files can then be used to improve document ranking, improve web crawling, determine tiers in a multi-tiered index, determine where to insert a document in a multi-tiered index, determine link weights, and update a search engine index.

Using network traffic logs for search engine index updates

Using network traffic logs for search enhancement

In one embodiment, a tree traversal sequentially processes every level of two or more binary trees until reaching a leaf node for each the binary trees. Processing each level of the binary trees comprises: determining whether each traversed node of the remaining binary trees at the level is a non-leaf node or a leaf node; for the traversed leaf nodes at the level, remove the binary trees to which the leaf nodes belong from the remaining binary trees; and iteratively evaluating the traversed non-leaf nodes at the level of every n remaining binary trees until the non-leaf nodes at the level of all of the remaining binary trees are evaluated, where n&#x2267;2 and evaluating the non-leaf nodes of each n binary trees comprises concurrently evaluating the n non-leaf nodes of the n binary trees to determine to which one of the two child nodes of each of the n non-leaf nodes to traverse.

Concurrent traversal of multiple binary trees

A method and system for maintaining a binary tree of pointers to a stream of data and for searching same. A novel binary tree is created by a search engine in which the nodes associated with strings in the data stream which are closer to the current data stream position are nearer the root of the tree than nodes associated with strings which are farther. As the current position in the stream is advanced, the search engine inserts a new node to the tree for that position as the root node. The tree is then restructured based on the relative value of the strings of each node while preserving the temporal order of the tree such that strings nearer the current position are associated with nodes which are closer to the root. The tree is ideal for searching data for LZ77-based data compression, since a single traversal of the tree returns the longest match length with the smallest offset.

Growing community of inventors

Toronto, Canada

Tomi Johan Poutanen