Abstract: A computer-implemented method of analysing transcript data, comprising: receiving a plurality of transcripts and a set of outcome data for each transcript, the outcome data associating each transcript with one or more outcomes; receiving a query transcript; processing the query transcript and each transcript within the plurality of transcripts; comparing, the processed data representing the query transcript with the processed data representing the plurality of transcripts, to identify a subset of the plurality of transcripts that meet a threshold similarity criterion with respect to the query transcript; and thereby determining a relationship between the query transcript and one or more outcomes.

Abstract: A computer-implemented method is provided for taking one or more actions relating to therapy, the method comprising: obtaining data comprising audio data relating to a therapy session between a therapist and one or more patients; extracting text data from the audio data to form a transcript; dividing the transcript into a plurality of utterances; using at least a first part of a deep learning model to assign a semantic representation to each of the plurality of utterances to produce a plurality of assigned utterances; compiling the plurality of assigned utterances to form a representation of the therapy session; using at least a second part of a deep learning model, and an input comprising the representation of the therapy session, to obtain an output predicting a characteristic of the therapist, and/or the therapy, and/or the one or more patient; and causing the system to take one or more actions relating to the therapy, wherein the one or more actions are selected based on the output meeting one or more pre

Abstract: Techniques are described for determining the value of individual facts in a knowledge base, and various applications of such fact values. In one example, the knowledge base is part of a question answering system. A ranking of knowledge base facts based on the number of times each of the knowledge base facts is used in answering user questions (e.g., as determined from question answering logs) is used to derive a fact value function that may then be used to determine the value of other facts included in or subsequently added to the knowledge base.

Abstract: A computer-implemented method is provided for taking one or more action relating to therapy. The method comprising: receiving text data relating to a therapy session between a therapist and a patient; dividing the text data into a plurality of utterances; assigning a semantic representation to each of the plurality of utterances to produce a plurality of assigned utterances; aggregating the plurality of assigned utterances to form a representation of the therapy session; providing an output prediction, based on the representation of the therapy session and optionally one or more further input, of one or more characteristic of at least one of the patient, the therapist and the therapy; taking one or more action relating to the therapy, wherein the one or more action is selected based on the output prediction meeting a predetermined criterion.

Abstract: A query parsing system uses a multi-stage process to parse the text of incoming queries before attempting to answer the queries. The multi-stage configuration involves a first trained classifier to determine the query type, or intent, of the text and a plurality of second trained classifiers, where each of the second trained classifiers is configured particularly for one specific respective query type. During query processing, the first trained classifier is used on the text to identify the query type. A second trained classifier for that specific identified query type is then found and used on the text to identify what strings in the text correspond to specific entities needed to resolve the query. The identified text strings and query type are then placed into a form understandable by a knowledge base and sent to the knowledge base for resolution. The classifiers may be trained using queries and answers previously processed by the knowledge base using a rules/template resolution process.

Abstract: A method of determining the effectiveness of a therapist, comprising: obtaining data relating to one or more patient variables and/or one or more service variables for one or more patient suffering from a mental health disorder and allocated to the therapist; attributing a score to the data for each of the patient variables and/or the service variables, wherein the scores are based on a correlation between historic cohort treatment outcomes and historic cohort data comprising cohort patient and/or service variables; combining the scores to calculate an aggregate score; comparing the aggregate score with a scale to produce a prediction of psychological therapy outcome for the one or more patient; obtaining an observation of psychological therapy outcome for the one or more patient after treatment by the therapist has been provided; and comparing the observation of psychological therapy outcome and the prediction of psychological therapy outcome for the one or more patient to make a determination of the effecti

Abstract: A computer-based psychotherapy triage method comprising: obtaining text data relating to a patient at an initial stage of a therapy process; using at least a first part of a deep learning model to obtain a representation of at least the text data; using at least a second part of the deep learning model, and an input thereto formed using the representation, to obtain an output predicting a characteristic of a condition of the patient and/or of the therapy process; and causing the system to take one or more actions relating to the therapy process, wherein the one or more actions are selected based on the output; wherein the deep learning model is trained using a training set comprising, for a plurality of other patients, text data relating to the other patient at an initial stage of a therapy process and a result of a determination of the characteristic.

Abstract: A voice-controlled question answering system that uses both a knowledge base and a detailed index of other sources that may be accessible over the Internet. The knowledge base is used to answer questions of more general interest whose answers are contained in the knowledge base. The index is used to answer more complex questions that are not answerable using the knowledge base. Web and other sources are analyzed to create the index, where text segments are indexed along with data describing the text segments in a quasi-semantic way. Quasi-semantic features are extracted from incoming spoken questions and used, along with machine learning trained models, to identify an indexed text segment that includes the answer to the question. The text segment may then be rearranged or simply passed to a speech synthesizer so the answer may be spoken aloud to a user.

Abstract: Techniques are described for adding knowledge to a knowledge base.