Abstract: An apparatus and a method for preserving privacy in natural language databases are provided. Natural language input may be received. At least one of sanitizing or anonymizing the natural language input may be performed to form a clean output. The clean output may be stored.

Abstract: A system is provided that trains a spoken language understanding (SLU) classifier. A corpus of user utterances is received. For each of the user utterances in the corpus, the user utterance is semantically parsed, and the result of this semantic parsing is represented as a rooted semantic parse graph. The parse graphs representing all of the user utterances in the corpus are then combined into a single corpus graph that represents the semantic parses of the entire corpus. The user utterances in the corpus are then clustered into intent-wise homogeneous groups of user utterances, where this clustering includes finding subgraphs in the corpus graph that represent different groups of user utterances, and each of these different groups has a similar user intent. The intent-wise homogeneous groups of user utterances are then used to train the SLU classifier, and the trained SLU classifier is output.

Abstract: Natural language query translation may be provided. A statistical model may be trained to detect domains according to a plurality of query click log data. Upon receiving a natural language query, the statistical model may be used to translate the natural language query into an action. The action may then be performed and at least one result associated with performing the action may be provided.

Abstract: Disclosed is a method and apparatus for responding to an inquiry from a client via a network. The method and apparatus receive the inquiry from a client via a network. Based on the inquiry, question-answer pairs retrieved from the network are analyzed to determine a response to the inquiry. The QA pairs are not predefined. As a result, the QA pairs have to be analyzed in order to determine whether they are responsive to a particular inquiry. Questions of the QA pairs may be repetitive and similar to one another even for very different subjects, and without additional contextual and meta-level information, are not useful in determining whether their corresponding answer responds to an inquiry.

Abstract: A relation detection model training solution. The relation detection model training solution mines freely available resources from the World Wide Web to train a relationship detection model for use during linguistic processing. The relation detection model training system searches the web for pairs of entities extracted from a knowledge graph that are connected by a specific relation. Performance is enhanced by clipping search snippets to extract patterns that connect the two entities in a dependency tree and refining the annotations of the relations according to other related entities in the knowledge graph. The relation detection model training solution scales to other domains and languages, pushing the burden from natural language semantic parsing to knowledge base population. The relation detection model training solution exhibits performance comparable to supervised solutions, which require design, collection, and manual labeling of natural language data.

Abstract: Natural language query translation may be provided. A statistical model may be trained to detect domains according to a plurality of query click log data. Upon receiving a natural language query, the statistical model may be used to translate the natural language query into an action. The action may then be performed and at least one result associated with performing the action may be provided.

Abstract: Systems and methods are provided for improving language models for speech recognition by personalizing knowledge sources utilized by the language models to specific users or user-population characteristics. A knowledge source, such as a knowledge graph, is personalized for a particular user by mapping entities or user actions from usage history for the user, such as query logs, to the knowledge source. The personalized knowledge source may be used to build a personal language model by training a language model with queries corresponding to entities or entity pairs that appear in usage history. In some embodiments, a personalized knowledge source for a specific user can be extended based on personalized knowledge sources of similar users.

Abstract: Click logs are automatically mined to assist in discovering candidate variations for named entities. The named entities may be obtained from one or more sources and include an initial list of named entities. A search may be performed within one or more search engines to determine common phrases that are used to identify the named entity in addition to the named entity initially included in the named entity list. Click logs associated with results of past searches are automatically mined to discover what phrases determined from the searches are candidate variations for the named entity. The candidate variations are scored to assist in determining the variations to include within an understanding model. The variations may also be used when delivering responses and displayed output in the SLU system. For example, instead of using the listed named entity, a popular and/or shortened name may be used by the system.

Abstract: Systems and methods for determining knowledge-guided information for a recurrent neural networks (RNN) to guide the RNN in semantic tagging of an input phrase are presented. A knowledge encoding module of a Knowledge-Guided Structural Attention Process (K-SAP) receives an input phrase and, in conjunction with additional sub-components or cooperative components generates a knowledge-guided vector that is provided with the input phrase to the RNN for linguistic semantic tagging. Generating the knowledge-guided vector comprises at least parsing the input phrase and generating a corresponding hierarchical linguistic structure comprising one or more discrete sub-structures. The sub-structures may be encoded into vectors along with attention weighting identifying those sub-structures that have greater importance in determining the semantic meaning of the input phrase.

Abstract: Systems for improving or generating a spoken language understanding system using a multitask learning method for intent or call-type classification. The multitask learning method aims at training tasks in parallel while using a shared representation. A computing device automatically re-uses the existing labeled data from various applications, which are similar but may have different call-types, intents or intent distributions to improve the performance. An automated intent mapping algorithm operates across applications. In one aspect, active learning is employed to selectively sample the data to be re-used.

Abstract: Disclosed herein is a system, method and computer readable medium storing instructions related to semantic and syntactic information in a language understanding system. The method embodiment of the invention is a method for classifying utterances during a natural language dialog between a human and a computing device. The method comprises receiving a user utterance; generating a semantic and syntactic graph associated with the received utterance, extracting all n-grams as features from the generated semantic and syntactic graph and classifying the utterance. Classifying the utterance may be performed any number of ways such as using the extracted n-grams, a syntactic and semantic graphs or writing rules.

Abstract: Functionality is described herein for determining the intents of linguistic items (such as queries), to produce intent output information. For some linguistic items, the functionality deterministically assigns intents to the linguistic items based on known intent labels, which, in turn, may be obtained or derived from a knowledge graph or other type of knowledge resource. For other linguistic items, the functionality infers the intents of the linguistic items based on selection log data (such as click log data provided by a search system). In some instances, the intent output information may reveal new intents that are not represented by the known intent labels. In one implementation, the functionality can use the intent output information to train a language understanding model.

Abstract: A processing unit can train a model as a joint multi-domain recurrent neural network (JRNN), such as a bi-directional recurrent neural network (bRNN) and/or a recurrent neural network with long-short term memory (RNN-LSTM) for spoken language understanding (SLU). The processing unit can use the trained model to, e.g., jointly model slot filling, intent determination, and domain classification. The joint multi-domain model described herein can estimate a complete semantic frame per query, and the joint multi-domain model enables multi-task deep learning leveraging the data from multiple domains. The joint multi-domain recurrent neural network (JRNN) can leverage semantic intents (such as, finding or identifying, e.g., a domain specific goal) and slots (such as, dates, times, locations, subjects, etc.) across multiple domains.

Abstract: A processing unit can extract salient semantics to model knowledge carryover, from one turn to the next, in multi-turn conversations. Architecture described herein can use the end-to-end memory networks to encode inputs, e.g., utterances, with intents and slots, which can be stored as embeddings in memory, and in decoding the architecture can exploit latent contextual information from memory, e.g., demographic context, visual context, semantic context, etc. e.g., via an attention model, to leverage previously stored semantics for semantic parsing, e.g., for joint intent prediction and slot tagging. In examples, architecture is configured to build an end-to-end memory network model for contextual, e.g., multi-turn, language understanding, to apply the end-to-end memory network model to multiple turns of conversational input; and to fill slots for output of contextual, e.g., multi-turn, language understanding of the conversational input.

Abstract: A device may facilitate a query dialog involving queries that successively modify a query state. However, fulfilling such queries in the context of possible query domains, query intents, and contextual meanings of query terms may be difficult. Presented herein are techniques for modifying a query state in view of a query by utilizing a set of query state modifications, each representing a modification of the query state possibly intended by the user while formulating the query (e.g., adding, substituting, or removing query terms; changing the query domain or query intent; and navigating within a hierarchy of saved query states). Upon receiving a query, an embodiment may calculate the probability of the query connoting each query state modification (e.g., using a Bayesian classifier), and parsing the query according to a query state modification having a high probability (e.g., mapping respective query terms to query slots within the current query intent).

Abstract: Systems and methods are provided for training language models using in-domain-like data collected automatically from one or more data sources. The data sources (such as text data or user-interactional data) are mined for specific types of data, including data related to style, content, and probability of relevance, which are then used for language model training. In one embodiment, a language model is trained from features extracted from a knowledge graph modified into a probabilistic graph, where entity popularities are represented and the popularity information is obtained from data sources related to the knowledge. Embodiments of language models trained from this data are particularly suitable for domain-specific conversational understanding tasks where natural language is used, such as user interaction with a game console or a personal assistant application on personal device.

Abstract: Utterance data that includes at least a small amount of manually transcribed data is provided. Automatic speech recognition is performed on ones of the utterance data not having a corresponding manual transcription to produce automatically transcribed utterances. A model is trained using all of the manually transcribed data and the automatically transcribed utterances. A predetermined number of utterances not having a corresponding manual transcription are intelligently selected and manually transcribed. Ones of the automatically transcribed data as well as ones having a corresponding manual transcription are labeled. In another aspect of the invention, audio data is mined from at least one source, and a language model is trained for call classification from the mined audio data to produce a language model.

Abstract: An apparatus and a method are provided for building a spoken language understanding model. Labeled data may be obtained for a target application. A new classification model may be formed for use with the target application by using the labeled data for adaptation of an existing classification model. In some implementations, the existing classification model may be used to determine the most informative examples to label.

Abstract: A method for assisting a user with one or more desired tasks is disclosed. For example, an executable, generic language understanding module and an executable, generic task reasoning module are provided for execution in the computer processing system. A set of run-time specifications is provided to the generic language understanding module and the generic task reasoning module, comprising one or more models specific to a domain. A language input is then received from a user, an intention of the user is determined with respect to one or more desired tasks, and the user is assisted with the one or more desired tasks, in accordance with the intention of the user.

Abstract: This disclosure pertains to a classification model, and to functionality for producing and applying the classification model. The classification model is configured to discriminate whether an input linguistic item (such as a query) corresponding to either a natural language (NL) linguistic item or a keyword language (KL) linguistic item. An NL linguistic item expresses an intent using a natural language, while a KL linguistic item expresses the intent using one or more keywords. In a training phase, the functionality produces the classification model based on query click log data or the like. In an application phase, the functionality may, among other uses, use the classification model to filter a subset of NL linguistic items from a larger set of items, and then use the subset of NL linguistic items to train a natural language interpretation model, such as a spoken language understanding model.