摘要:
At least some embodiments are directed to a system to compute uniform structured summarization of customer chats. In at least some embodiments, the system may operate a processor and receive a corpus of chats between customers and customer service representatives of an enterprise. Grouping the corpus of chats into subgroup task types and then extracting chat keywords and chat related words for each subgroup task type. Generating an expandable template data structure for each subgroup task type. Processing at least one chat to extract chat utterances and chat snippets ranking the chat utterances and chat snippets. Populating the expandable template data structure based on rankings to generate a chat summary data structure.
摘要:
Disclosed is a content publisher recommender system comprising: a server; a content database, configured within the server, within which are stored one or more channels associated with an end user, wherein the one or more channels stores at least one end user identification and a user historical database including a time log of the end user's activity, a log of search phrases used by the end user, a record of results identified on the search phrases; a ranking of the record of results, a record of which results the end user interacts with and a record of the format of the first clicked results from an end user wherein the search phrases are analysed using natural language processing to reduce the terms of the search phrases to their root form, the terms of the phrase are broken down into their part of speech and all keywords identified during natural language processing have stop words removed; and wherein considerations of the stored data and the analysis provides a ranking for searching to provide recommended content.
摘要:
At least some embodiments are directed to a system to compute uniform structured summarization of customer chats. In at least some embodiments, the system may operate a processor and receive a corpus of chats between customers and customer service representatives of an enterprise. Grouping the corpus of chats into subgroup task types and then extracting chat keywords and chat related words for each subgroup task type. Generating an expandable template data structure for each subgroup task type. Processing at least one chat to extract chat utterances and chat snippets ranking the chat utterances and chat snippets. Populating the expandable template data structure based on rankings to generate a chat summary data structure.
摘要:
At least some embodiments are directed to a system to compute uniform structured summarization of customer chats. In at least some embodiments, the system may operate a processor and receive a corpus of chats between customers and customer service representatives of an enterprise. Grouping the corpus of chats into subgroup task types and then extracting chat keywords and chat related words for each subgroup task type. Generating an expandable template data structures for each subgroup task type. Processing at least one chat to extract chat utterances and chat snippets ranking the chat utterances and chat snippets. Populating the expandable template data structure based on rankings to generate a chat summary data structure.
摘要:
A method for managing semantic information on an M2M/IoT platform is provided. The method for managing semantic information according to an embodiment of the present invention stores semantic data in the first attribute of an M2M resource and updates a part of the semantic data stored in the first attribute. Accordingly, efficient management of semantic information on an M2M/IoT platform is possible, and particularly a partial update of the semantic information can be performed.
摘要:
Techniques for processing a speech to text query are described herein. The techniques may include receiving a plurality of speech to text translation alternatives for a phrase of a natural language query, and tagging and parsing each of the translation alternatives based on a static analysis of the known domain that is at least partially structured, known tags of the known domain, and custom rules. The techniques may also include ranking the translation alternatives based on the tagging and parsing and translating the phrase based on the ranking.
摘要:
A capitalization-sensitive (“Cap-sensitive”) search is a feature that provides the ability to specify as search criteria both a term and a capitalization characteristic or signature of the term. There are conventional approaches that enable a capital-sensitive approach to be performed from within a document.
摘要:
For automated text processing, the inventors devised, among other things, an exemplary system that includes an entity tagger, an entity resolver, a text segment classifier, and a relationship extractor. The entity tagger receives an input text segment, and tags named entities with the segment as being a person, company, or place. The entity resolver accesses authority files, and associates the persons and companies named in the text segment with specific entries in the files. The text segment classifier determines whether the text segment includes a relationship event, such as job-change event or merger and acquisition event, and if an event is detected, the relationship extractor determines the event role of entities named in the segment. For example, the extractor determines for a merger and acquisition event, which named company was the acquirer and which was acquired.
摘要:
Predicting potential policy violations in a given data source (such as a set of documents) so that a deeper analysis can be performed on the set of documents to gain additional insights into the potential policy violation(s) that may be contained therein. In some instances, this prediction is done by performing a random sampling operation on the set of documents and collecting the metadata from these documents in order to determine whether or not the set of documents that are randomly sampled include a policy violation that can be remediated.
摘要:
Systems and methods for providing dynamic responses to a user query are disclosed herein. A system receives a message request from a computing device, and generates an intent associated with the message request based on a pre-processing of the message request. Further, the system concurrently transmits a query to a database for a prestored response to the message request based on the generated intent, and the generated intent to a knowledge source searcher. The knowledge source searcher asynchronously queries one or more knowledge sources for a predicted response to the message request. Furthermore, the system receives the prestored response from the database and the predicted response from the knowledge source searcher, determines a response to the message request, from at least one of the prestored response and the predicted response, based on a scoring model, and transmits the determined response to the computing device.