Abstract: Methods and apparatuses are described for determining future user actions using time-based featurization of clickstream data. A server captures clickstream data corresponding to web browsing sessions and converts the clickstream data into tokens by identifying each unique URL and parsing each unique URL into tokens. The server generates a frequency matrix based upon the tokens, and generates a latent feature vector for each URL in the session based upon the frequency matrix. The server merges the latent feature vectors and the clickstream data into an aggregate clickstream vector set for a user. The server assigns time-decayed weight values to each latent feature vector in the aggregate clickstream vector set. The server combines the time-decayed latent feature vectors to generate a clickstream embedding for the user, and executes a machine learning model using the clickstream embedding to generate one or more predicted actions of the user.

Abstract: Methods and apparatuses are described for normalizing digital content across databases and generating personalized content recommendations. A server normalizes structured text for each content item to generate unstructured text. The server converts the unstructured text into a multidimensional content item feature set. The server trains a model based upon user profile information, historical content consumption information, historical content recommendation information, and the feature sets. The server receives a request including a vector associated with a user of a client device. The server executes the model using the vector as input to generate interaction prediction scores. The server selects scores above a threshold and identifies content items associated with each score.

Abstract: Methods and apparatuses are described for digital content classification and recommendation based upon reinforcement learning. A server converts unstructured text corresponding to each digital content item into a content item feature set. The server generates a user context vector associated with a plurality of users. The server trains a linear multi-armed bandit (MAB) classification model based upon the user context vectors and historical user content recommendation information. The server receives a new user context vector associated with a new user. The server executes the MAB model using the new user context vector as input to generate content interaction prediction scores. The server selects the content interaction prediction scores above a predetermined threshold and identifies the associated digital content item. The server presents the identified digital content items on a client device and receives a response. The server updates linear UCB coefficient vectors of the MAB model based upon the response.

Abstract: Methods and apparatuses are described for normalizing digital content across databases and generating personalized content recommendations. A server normalizes structured text for each content item to generate unstructured text. The server converts the unstructured text into a multidimensional content item feature set. The server trains a model based upon user profile information, historical content consumption information, historical content recommendation information, and the feature sets. The server receives a request including a vector associated with a user of a client device. The server executes the model using the vector as input to generate interaction prediction scores. The server selects scores above a threshold and identifies content items associated with each score.

Abstract: Methods and apparatuses are described for digital content classification and recommendation based upon reinforcement learning. A server converts unstructured text corresponding to each digital content item into a content item feature set. The server generates a user context vector associated with a plurality of users. The server trains a linear multi-armed bandit (MAB) classification model based upon the user context vectors and historical user content recommendation information. The server receives a new user context vector associated with a new user. The server executes the MAB model using the new user context vector as input to generate content interaction prediction scores. The server selects the content interaction prediction scores above a predetermined threshold and identifies the associated digital content item. The server presents the identified digital content items on a client device and receives a response. The server updates linear UCB coefficient vectors of the MAB model based upon the response.

Abstract: Methods and apparatuses are described for automated predictive product recommendations using reinforcement learning. A server captures historical activity data associated with a plurality of users. The server generates a context vector for each user, the context vector comprising a multidimensional array corresponding to historical activity data. The server transforms each context vector into a context embedding. The server assigns each context embedding to an embedding cluster. The server determines, for each context embedding, (i) an overall likelihood of successful attempt and (ii) an incremental likelihood of success associated products available for recommendation. The server calculates, for each context embedding, an incremental income value associated with each of the likelihoods of success. The server aggregates (i) the overall likelihood of successful attempt, (ii) the likelihoods of success, and (iii) the incremental income values into a recommendation matrix.

Abstract: Methods and apparatuses are described for automated predictive call routing using reinforcement learning. A server captures a bitstream of an incoming call from a first client device, the bitstream including metadata comprising attributes of the incoming call and attributes of a user of the device. The server determines an identity of the user based upon the metadata. The server generates a first context vector comprising a multidimensional array corresponding to the metadata. The server inserts the first vector into a high-dimensional vector space comprising historical context vectors, each historical vector (i) corresponding to metadata associated with a historical call and (ii) associated with an income value and a routing decision. The server determines historical vectors in proximity to the first vector. The server identifies one of the determined historical vectors with an optimal income value and routes the bitstream to a second device using the routing decision.

Abstract: Methods and apparatuses are described for automated routing of voice calls using time-based predictive clickstream data. A server captures clickstream data comprising uniform resource locators (URLs) and one or more timestamps of a web session. The server converts the clickstream data into tokens and generates a frequency matrix based upon the tokens. The server generates a feature vector based upon the frequency matrix. The server receives an incoming voice call from a remote device and identifies that the remote device is associated with a user of the client computing device. The server determines intent for the incoming voice call based upon the feature vector, and routes the incoming voice call to a destination device based upon the determined intent.