Abstract: Certain aspects produce a scoring model that can automatically classify future text samples. In some examples, a processing device perform operations for producing a scoring model using active learning. The operations includes receiving existing text samples and searching a stored, pre-trained corpus defining embedding vectors for selected words, phrases, or documents to produce nearest neighbor vectors for each embedding vector. Nearest neighbor selections are identified based on distance between each nearest neighbor vector and the embedding vector for each selection to produce a text cloud. Text samples are selected from the text cloud to produce seed data that is used to train a text classifier. A scoring model can be produced based on the text classifier. The scoring model can receive a plurality of new text samples and provide a score indicative of a likelihood of being a member of a selected class.

Abstract: Certain aspects produce a scoring model that can automatically classify future text samples. In some examples, a processing device perform operations for producing a scoring model using active learning. The operations includes receiving existing text samples and searching a stored, pre-trained corpus defining embedding vectors for selected words, phrases, or documents to produce nearest neighbor vectors for each embedding vector. Nearest neighbor selections are identified based on distance between each nearest neighbor vector and the embedding vector for each selection to produce a text cloud. Text samples are selected from the text cloud to produce seed data that is used to train a text classifier. A scoring model can be produced based on the text classifier. The scoring model can receive a plurality of new text samples and provide a score indicative of a likelihood of being a member of a selected class.

Abstract: A message retrieval subsystem comprises a voice extensible mark-up language browser, a message server, and a message library. The VXML browser is configured with an application programming interface that enables a subscriber of a distributed voice messaging service to request a stored message. The message server receives a request to open a connection from a client process operative on the voice extensible mark-up language browser, spawns a thread to enable the connection, uses the thread to queue one or more subscriber requests for respective blocks of audio information, removes and processes the one or more subscriber requests, and uses the thread to forward a retrieved block of audio information to a subscriber. The message library is coupled to the message server and interposed between the message server and a common message store. The upper library isolates the VXML browser from a streaming protocol used to access the stored message.

Abstract: A message retrieval subsystem comprises a voice extensible mark-up language browser, a message server, and a message library. The VXML browser is configured with an application programming interface that enables a subscriber of a distributed voice messaging service to request a stored message. The message server receives a request to open a connection from a client process operative on the voice extensible mark-up language browser, spawns a thread to enable the connection, uses the thread to queue one or more subscriber requests for respective blocks of audio information, removes and processes the one or more subscriber requests, and uses the thread to forward a retrieved block of audio information to a subscriber. The message library is coupled to the message server and interposed between the message server and a common message store. The upper library isolates the VXML browser from a streaming protocol used to access the stored message.