Abstract: Embodiments are directed to a method of utilizing an ensemble of distributional semantics systems in conjunction with a domain term extractor for generating domain-specific synonyms. The method allows for extraction of high-quality, domain-specific synonyms that can be used in an information handling system, such as a question-answer system or in an information retrieval (IR) system, capable of processing natural language. According to embodiments, the domain term extractor identifies the words for which synonyms are sought, and the ensemble of distributional semantics systems determines the synonyms.

Abstract: A technique for resolving entities provided in a question includes creating respective entity context vectors (ECVs) for respective entities in an applicable knowledge graph (KG). A question is received from a user. A first entity is identified in the question. The first entity is associated with a matching one of the entities in the KG. An ECV for the matching one of the entities in the KG is modified. An answer to the question is generated based on the modified ECV.

Abstract: Embodiments provide a computer implemented method for generating a domain-specific thesaurus on a cognitive system, comprising: receiving data of the domain-specific corpus and a plurality of terms of interest from a user; splitting the data of the domain-specific corpus into a plurality of sentences using natural language processing techniques; for each term in the plurality of terms of interest, retrieving a plurality of candidate sentences containing a corresponding term, from the plurality of sentences; for each candidate sentence, providing a list of synonyms of the corresponding term, wherein the synonyms are contextual alternatives in the corresponding candidate sentence; for each term in the plurality of terms of interest, tracking a frequency of each synonym, and forming a frequency map including all the synonyms of a corresponding term and the frequency of each synonym; and generating a domain-specific thesaurus based on a combination of all the synonyms in the frequency map.

Abstract: An approach is provided that receives a message and applies a deep analytic analysis to the message. The deep analytic analysis results in a set of enriched message embedding (EME) data that is passed to a trained neural network. Based on a set of scores received from the trained neural network, a conversation is identified from a number of available conversations to which the received message belongs. The received first message is then associated with the identified conversation.

Abstract: An approach is provided that receives a message and applies a deep analytic analysis to the message. The deep analytic analysis results in a set of enriched message embedding (EME) data that is passed to a trained neural network. Based on a set of scores received from the trained neural network, a conversation is identified from a number of available conversations to which the received message belongs. The received first message is then associated with the identified conversation.

Abstract: An approach is provided for automatically predicting an event occurrence based on a question from an end user presented using a near-real-time natural language processing (NLP) analysis to generate, score and rank a plurality of event occurrences based on a plurality of question context parameters extracted from the question, one or more user profile parameters for the end user, and the one or more historical questions, answers, and events having a specified spatial and/or temporal proximity to the question which are identified by an information handling system. In the approach, performed by an information handling system, a top ranked event occurrence from the ranked plurality of event occurrences is selected for inclusion in a notification message that is communicated or broadcast to the end user, as well as other users engaged with the information handling system and/or first responders in the affected area.

Abstract: In response to running at least one testing phrase on a previously trained text classifier and identifying a separate predicted classification label based on a score calculated for each respective at least one testing phrase, a text classifier decomposes extracted features summed in the score into word-level scores for each word in the at least one testing phrase. The text classifier assigns a separate heatmap value to each of the word-level scores, each respective separate heatmap value reflecting a weight of each word-level score. The text classifier outputs the separate predicted classification label and each separate heatmap value reflecting the weight of each word-level score for defining a heatmap identifying the contribution of each word in the at least one testing phrase to the separate predicted classification label for facilitating client evaluation of text classification anomalies.

Abstract: A test controller submits testing phrases to a text classifier and receives, from the text classifier, classification labels each comprising one or more respective heatmap values each associated with a separate word. The test controller aligns each of the classification labels corresponding with a respective testing phrase. The test controller identifies one or more anomalies of a selection of one or more classification labels that are different from an expected classification label for the respective testing phrase. The test controller outputs a graphical representation in a user interface of the selection of one or more classification labels and one or more respective testing phrases with visual indicators based on one or more respective heatmap values.

Abstract: A test controller submits testing phrases to a text classifier and receives, from the text classifier, classification labels each comprising one or more respective heatmap values each associated with a separate word. The test controller aligns each of the classification labels corresponding with a respective testing phrase. The test controller identifies one or more anomalies of a selection of one or more classification labels that are different from an expected classification label for the respective testing phrase. The test controller outputs a graphical representation in a user interface of the selection of one or more classification labels and one or more respective testing phrases with visual indicators based on one or more respective heatmap values.

Abstract: A system and a computer program product are provided for evaluating question-answer pairs in an answer key by comparing a first answer key answer to a plurality of candidate answers to determine if the answer key may have a problem if the plurality of candidate answers are more similar to one another than to the first answer and to determine if the plurality of candidate answers has gradient information which may be used to update the answer key if not already included in the answer key.

Abstract: A method, system and computer-usable medium are disclosed for preserving temporal relevance in a response to a query. A query is processed to extract first temporal features, which is then used to identify first documents within a corpus. The first documents are processed to generate first metadata and mined content, which is processed with the first temporal features to generate second documents having second temporal features corresponding to the first temporal features. The corpus is updated with new documents and revisions to the first documents to generate second documents. In turn, the second documents are processed to generate second metadata and mined content, which is then used to generate a ranked list of temporally-relevant documents.

Abstract: A computer-implemented method includes creating a classifier by: training a machine learning model using two or more tasks, wherein the tasks lie in two or more domains; including in the machine learning model at least one attribute common to at least two of said two or more domains; including in the machine learning model at least one latent feature that affects at least two of the two or more tasks that fall within one of the at least two domains; and constructing the classifier based on said machine learning model. The computer-implemented method further includes applying the classifier to at least one operational task.

Abstract: A computer-implemented method includes creating a classifier by: training a machine learning model using two or more tasks, wherein the tasks lie in two or more domains; including in the machine learning model at least one attribute common to at least two of said two or more domains; including in the machine learning model at least one latent feature that affects at least two of the two or more tasks that fall within one of the at least two domains; and constructing the classifier based on said machine learning model. The computer-implemented method further includes applying the classifier to at least one operational task.

Abstract: A natural language query (NLQ) is translated to a structured data query (e.g., a SQL statement) by extracting entities from the NLQ and replacing them with generic variables to form a generic query. The generic query is associated with a structured question type which includes structured data variables using natural language classifiers (NLCs). Specific data is inserted in the structured question type in relation to the structured data variables based on the extracted entities to form the structured data query. An ensemble of NLCs trained with different ground truths can be used to yield multiple candidate question types. One of the candidate question types is selected based on confidence levels. The multiple NLCs can include an NLC which is optimized according to a focus of the generic query. For example, an NLC can be optimized for a specific data structure (such as SQL), or for comparative queries.

Abstract: Natural language processing (NLP) with awareness of textual polarity. An NLP system, such as a search engine or a Question-Answering (QA) system receives input text for processing. The input text may be a text fragment, a search phrase, a question having a general type, or a polar question having a yes or no answer. The NLP system identifies textual polarity and provides responses to the input text (for example, in answer form) based on identifying evidence whose selection, scoring, and processing, is informed by the textual polarity of the input text, and the textual polarity of candidate evidence passages.

Abstract: Natural language processing (NLP) with awareness of textual polarity. An NLP system, such as a search engine or a Question-Answering (QA) system receives input text for processing. The input text may be a text fragment, a search phrase, a question having a general type, or a polar question having a yes or no answer. The NLP system identifies textual polarity and provides responses to the input text (for example, in answer form) based on identifying evidence whose selection, scoring, and processing, is informed by the textual polarity of the input text, and the textual polarity of candidate evidence passages.

Abstract: An approach is provided for automatically predicting an event occurrence based on a question from an end user presented using a near-real-time natural language processing (NLP) analysis to generate, score and rank a plurality of event occurrences based on a plurality of question context parameters extracted from the question, one or more user profile parameters for the end user, and the one or more historical questions, answers, and events having a specified spatial and/or temporal proximity to the question which are identified by an information handling system. In the approach, performed by an information handling system, a top ranked event occurrence from the ranked plurality of event occurrences is selected for inclusion in a notification message that is communicated or broadcast to the end user, as well as other users engaged with the information handling system and/or first responders in the affected area.

Abstract: An approach is provided for identifying entity relationships based on word classifications extracted from business documents stored in a plurality of corpora. In the approach, performed by an information handling system, a plurality of cluster classifications are identified for the business documents so that entity information from the business documents can be classified or assigned to the cluster classifications, such as by performing natural language processing (NLP) analysis of the business documents. The approach applies semantic analysis to identify and score entity relationships between the entity information classified in the cluster classifications, and based on the scored entity relationships, cluster relationships between the cluster classifications are identified.

Abstract: Electronic natural language processing in a natural language processing (NLP) system, such as a Question-Answering (QA) system. A receives electronic text input, in question form, and determines a readability level indicator in the question. The readability level indicator includes at least a grammatical error, a slang term, and a misspelling type. The computer determines a readability level for the electronic text input based on the readability level indicator, and retrieves candidate answers based on the readability level.

Abstract: A method, in a data processing system comprising a processor and a memory, for analyzing product impact, the method comprising receiving, by the data processing system, data representative of an entity's operations, receiving, by the data processing system, a selection of one or more products, parsing, by the data processing system, features of the one or more products from product data input, and generating, by the data processing system, product profiles for the one or more products based on the parsed features. The method further comprising generating, by the data processing system, risk assessment data based on the data representative of the entity's operations and the product profiles, the risk assessment data including an indication of one or more changes in risk metrics of one or more risk components associated with the entity's operations by offering the one or more products.