Robert Boschgmbh

Robert Bosch Gnbh

Weng, Fuliang

Feng, Zhe

Hasan, Taufiq

Zhe Feng

Fuliang Weng

Lin Zhao

Kui Xu

Zhongnan Shen

Haibo Ding

Baoshi Yan

Shabnam Ghaffarzadegan

Chin-Hui Lee

Taufiq Hasan

Yifan He

Zhengyu Zhou

Bingqing Wang

You-Chi Cheng

Katrina Li

Xinyan Zhao

Liu Ren

Yu Wang

Doo Soon Kim

Lincan Zou

Feng Lin

Badri Raghunathan

Jun Araki

Yiwei Sun

Fei Liu

Qi Zhang

Attila Reiss

Junling Hu

Mirko Ruhs

Tobias Scheideck

Robert Duerichen

Tobias Menne

Jiacheng Li

Kehuang Li

Asif Salekin

Ishan Dindorkar

Lars Franke

Laura Stoia

Junkuo Cao

William Chang

Linfeng Song

Disclosed are systems and methods for a computerized framework that provides a document structure parsing system for requirement engineering documents, where the logical structure of the text is not available, and is to be rebuilt based on the raw textual content. The framework approaches the build of the logical structure according to two phases. The first phase involves creating a list of list of text snippets from the raw text, where sequence labeling is adopted to re-segment and merge initially segmented text snippets. The second phase involves the framework executing computerized techniques including embedding adaptation approach, a hierarchy structure rebuilt algorithm, and a requirement text selection strategy to rebuild the hierarchy structure.

Computerized systems and methods for hierarchical structure parsing and building

A computer-implemented system and method relate to natural language processing. A candidate is selected based on a query. The candidate includes a question and an answer to that question. The candidate is classified as belonging to a class. The class indicates that the candidate is relevant to the query. The class is selected from among a group of classes, which includes at least that class and another class. The another class indicates that the candidate has the answer to the query. Upon classifying the candidate as belonging to the class, a machine learning system is configured to generate a relevant response for the query upon receiving the query and the question as input data. The relevant response provides information associated with a keyword of the question that is not contained in the query.

Retrieval-based dialogue system with relevant responses

An automatic terminology linking system includes a candidate generator configured to identify candidate nodes for each terminology that is to be linked to a node of the knowledge base. A pseudo-candidate generator is configured to identify pseudo-candidate nodes for candidate-less terminologies. A candidate scorer is configured to respectively score the candidate nodes and the pseudo-candidate nodes by collective inference using occurrence statistics and co-occurrence statistics for these nodes. The pseudo-candidate generator is configured to identify knowledge base nodes that are semantically-related to candidate-less terminology as the pseudo-candidate nodes for the candidate-less terminology.

System and method for domain-independent terminology linking

A method and system for detecting and localizing a target audio event in an audio clip is disclosed. The method and system use utilizes a hierarchical approach in which a dilated convolutional neural network to detect the presence of the target audio event anywhere in an audio clip based on high level audio features. If the target audio event is detected somewhere in the audio clip, the method and system further utilizes a robust audio vector representation that encodes the inherent state of the audio as well as a learned relationship between state of the audio and the particular target audio event that was detected in the audio clip. A bi-directional long short term memory classifier is used to model long term dependencies and determine the boundaries in time of the target audio event within the audio clip based on the audio vector representations.

System and method for audio event detection in surveillance systems

A method of automatically generating a terminology definition knowledge base (KB) includes mapping each word in a word sequence to a real value dense vector using dense vector representations. The word sequence is then processed using a Convolutional Neural Network (CNN) model to identify whether the word sequence includes a terminology definition and to label the word sequence with a label indicating whether a terminology definition exists within the word sequence. The word sequence is then processed using a Conditional Random Field (CRF) model to identify boundaries of the terminology definition in the word sequence. The terminology definition is then extracted and added to the terminology definition KB.

System and method for domain- and language-independent definition extraction using deep neural networks

Systems and methods for weakly-supervised training a machine-learning model to perform named-entity recognition. All possible entity candidates and all possible rule candidates are automatically identified in an input data set of unlabeled text. An initial training of the machine-learning model is performed using labels assigned to entity candidates by a set of seeding rules as a first set of training data. The trained machine-learning model is then applied to the unlabeled text and a subset of rules from the rule candidates is identified that produces labels that most accurately match the labels assigned by the trained machine-learning model. The machine-learning model is then retrained using the labels assigned by the identified subset of rules as the second set of training data. This process is iteratively repeated to further refine and improve the performance of the machine-learning model for named-entity recognition.

Weakly supervised and explainable training of a machine-learning-based named-entity recognition (NER) mechanism

A server includes a controller in communication with the server and configured to receive a query related to an issue with a vehicle, retrieve information related to the query from a database that includes structured data including data related to vehicle issues and a solution to the vehicle issues, determine whether the database includes structured data including data related to the solution responsive to the query, and output information including the solution to the issue when the determination identifies that the structured data including the solution in the database or retrieve information from a website when the determination identifies that the structured data including the solution is not found in the database, wherein the controller is further configured to retrieve the information from the website by conducting a keyword search for a first set of results and filtering the information from the website utilizing an autoencoder.

System and method for information extraction and retrieval for automotive repair assistance

A system for automatically labeling data using conceptual descriptions. In one example, the system includes an electronic processor configured to generate unlabeled training data examples from one or more natural language documents and, for each of a plurality of categories, determine one or more concepts associated with a conceptual description of the category and generate a weak annotator for each of the one or more concepts. The electronic processor is also configured to apply each weak annotator to each training data example and, when a training data example satisfies a weak annotator, output a category associated with the weak annotator. For each training data example, the electronic processor determines a probabilistic distribution of the plurality of categories. For each training data example, the electronic processor labels the training data example with a category having the highest value in the probabilistic distribution determined for the training data example.

Automatically labeling data using conceptual descriptions

Systems and methods for training a machine-learning model for named-entity recognition. A rule graph is constructed including a plurality of nodes each corresponding to a different labeling rule of a set of labeling rules (including a set of seeding rules of known labeling accuracy and a plurality of candidate rules of unknown labeling accuracy). The nodes are coupled to other nodes based on which rules exhibit the highest sematic similarity. A labeling accuracy metric is estimated for each candidate rule by propagating a labeling confidence metric through the rule graph from the seeding rules to each candidate rule. A subset of labeling rules is then identified by ranking the rules by their labeling confidence metric. The identified subset of labeling rules is applied to unlabeled data to generate a set of weakly labeled named entities and the machine-learning model is trained based on the set of weakly labeled named entities.

Graph-based labeling rule augmentation for weakly supervised training of machine-learning-based named entity recognition

A method of detecting occupancy in an area includes obtaining, with a processor, an audio sample from an audio sensor and determining, with the processor, feature functional values of a set of selected feature functionals from the audio sample. The determining of the feature functional values includes extracting features in the set of selected feature functionals from the audio sample, and determining the feature functional values of the set of selected features from the extracted features. The method further includes determining, with the processor, occupancy in the area using a classifier based on the determined feature functional values.

System and method for determining occupancy

Methods and systems for performing semantic entity recognition. The method includes accessing a document stored in a memory and selecting from a general knowledge data repository, target domain information based on a specified target domain. The method also includes generating a plurality of weak annotators for the document based upon the selected target domain information and expert knowledge from a domain-specific expert knowledge data repository and applying the plurality of weak annotators to the document to generate a plurality of weak labels. The method further includes selecting at least one weak label from the plurality of weak labels as training data and training a semantic entity prediction model using the training data.

Weakly supervised semantic entity recognition using general and target domain knowledge

A system for ontology-aware sound classification. The system includes an electronic processor that is configured to create a first graph based on relationships between fine audio classification labels and create a second graph based on relationships between coarse audio classification labels. The electronic processor is also configured to receive an audio clip including one or more sounds, execute a first graph convolutional network with the first graph as input, and execute a second graph convolutional network with the second graph as input. Using the outputs of the first graph convolutional network and the second graph convolutional network, the electronic processor is configured to determine one or more coarse labels, one or more fine labels, or both to classify the one or more sounds in the audio clip.

Ontology-aware sound classification

DIY (Do-It-Yourself) is challenging for many novices, requiring extensive knowledge such as the usage of particular tools and the properties of the required materials. Many DIYers use web searches to find relevant information and instructions, but web search is time-consuming and the results often do not fit the DIYers' specific needs. To address these problems, we present a Question Answering (QA) system that can assist DIYers through the whole cycle of a DIY project. Given a natural language question about a DIY project, the QA system described herein provides an answer along with the explanations that are tailored to the DIYers' specific needs.

Knowledge-based question answering system for the DIY domain

A method for slot-filling in an automated spoken language understanding (SLU) system includes receiving words in a text phrase, generating an encoded output and attention weights using a recurrent neural network (RNN) encoder with an attention mechanism, generating a first and second probability distributions of slot words in an extended slot vocabulary database using an RNN decoder and a pointer network, respectively, for a time step, generating a combined probability distribution based on the first and second probability distributions, and identifying a slot word corresponding to one word in the extended slot vocabulary database having a highest probability value in the combined probability distribution for the time step.

Slot filling in spoken language understanding with joint pointer and attention

A system for monitoring dietary activity of a user includes a wearable device having at least one audio input unit configured to record an audio sample corresponding to audio from a user's neck. The system further includes a processor configured to execute programmed instructions stored in a memory to obtain an audio sample from the audio input unit of a wearable device, determine segmental feature values of a set of selected features from the audio sample by extracting short-term features in the set of selected features from the audio sample and determining the segmental feature values of the set of selected features from the extracted short-term features. The processor is further configured to, using a classifier, classify a dietary activity based on the determined segmental feature values of the audio sample and generate an output corresponding to the classified dietary activity.

System and method for monitoring dietary activity

Growing community of inventors

Mountain View, CA, United States of America

Zhe Feng