International Business Machines Corporation

Terrance Licensing LLC

Chih-Hsiung Liu

Joey H Y Tseng

Hao-Ting Shih

Wei-Ting Chou

Xin Peng Liu

Cheng-Fang Lin

I-Chien Lin

Peter Wu

Po-cheng Chiu

Norman Kung

Chen Zhang

June-Ray Lin

Chih-Wen Su

Tsai-Hsuan Hsieh

Yi-Hong Wang

Yu Chen Zhou

Wei Ey Li

Cheng-Ta Lee

Jin Zhang

Yang Wang

Shaw-Ben Shi

Pei Jian Liu

Gary Paul Noble

Guo Qiang Li

Bing Xin Wang

Neal J Alewine

Srinivasan Muralidharan

Ching-Wei Cheng

Ci-Wei Lan

Barry Mosakowski

Chiwen Chang

Kuo-Liang Chou

Min-Tsung Wu

Yu-chun Lin

Tzuching Kuo

Wan-Tseng Hsiao

Kate Lin

Tzu-Chen Chao

Hsin-Yu Hsieh

Cheng-Lun Yang

Li-Feng Tseng

Chen-Hsi Tsai

Ye Yu

Li Qiang Mx Xue

Jian Gang J G Jin

Wei E Y Li

Shen-Jai Liao

Li Qiang M X Xue

John C Bellessa

Yi-Ching Peng

Ming-Yu Wei

Chin-Teng Cheng

Ying-Hsien Lee

Wayen W J Wu

Jarwow C H Wu

Kevin C M Lai

Liang-Shu Huang

Joanne Kuei-Chen Ho

Jian Gang Jg Jin

Systems and methods for natural language generation by an edge computing device are disclosed. In one embodiments, a method comprises: receiving, by an edge computing device, event data from an edge event; determining, by the edge computing device, that a network connection to a cloud server is not available; extracting, by the edge computing device, features of the event data; predicting, by a local neural network of the edge computing device, an action for the edge computing device to take based on the features of the event data, wherein the action is associated with a confidence level; and determining, by the edge computing device, whether the confidence level meets a predetermined threshold value.

Natural language generation by an edge computing device

A method includes receiving a natural language query at an information system, the natural language query indicating an intent and at least a first factor and a second factor. The method also includes retrieving a set of candidate information from the information system based on the natural language query, the set of candidate information having a type determined by the intent. The method additionally includes selecting a knowledge display template from a set of knowledge display templates using the intent, the first factor and the second factor. The method further includes rendering, using the knowledge display template, a first knowledge graph comprising the set of candidate information, the first knowledge graph indicating a relationship between the set of candidate information based the first factor and the second factor.

Interactive information retrieval using knowledge graphs

An example operation may include one or more of configuring a blockchain network comprising first and second blockchain nodes, providing, by the first blockchain node, a data reference to the second blockchain node, accessing a document, by the second blockchain node, from the first blockchain node, and providing by the second blockchain node, a proof of receipt for the document to a shared blockchain ledger.

Document transfer processing for blockchains

Enhancing a data query is provided. A knowledge base is searched for information corresponding to the data query. It is determined whether a query result is greater than a query result threshold level of information for the data query based on an identified query intent and keywords of the data query. In response to determining that the query result is not greater than the query result threshold level of information for the data query, a set of questions is sent to a personal digital agent of a client device that sent the data query for additional information based on the identified query intent and keywords. Analytic concepts answering the set of questions are received from analytic modules of the personal digital agent that are based on IoT-based personal data corresponding to a user of the client device. The data query is enhanced using the analytic concepts.

Intelligent and interactive knowledge system

A single Internet of Things (IoT) gateway flow computer (either on a gateway machine or a non-gateway machine) that controls flow through both of the following types of gateways: (i) cloud gateways; and (ii) edge gateways. Both overall configuration and sub-configuration are automatically and dynamically controlled by the single, system-wide IoT gateway flow computer.

Internet of things (IoT) gateway flow engine optimizer and configuration distribution

A method is provided to enhance efficiency of sensor event data transmission over network. Specifically, a method is described to buffer a set of sensor data, to group one or more of the set of sensor data having a same type for batch processing. The batch processing includes compressing and securing operations on the grouped sensor data and restoring the original message sequence of the grouped sensor data.

Efficient sensor data delivery

An approach is provided for domain-specific chatbots that employ distributed natural language classifiers (NLCs). A NLC included in a first chatbot is executed, which generates a first intent of a question and a first confidence that the first intent is an actual intent of the question. Based on a determination that the first confidence does not exceed a threshold, the question is broadcasted to other chatbots. A second confidence that a second intent is an actual intent is received from a second chatbot in response to the broadcast, and is determined to exceed the threshold. An association among the question, the second chatbot, the second confidence, and the second intent is stored in a memory cache included in the first chatbot. A response to the question is generated based on the second intent. The response is presented to a user from the first chatbot.

Building a gossip group of domain-specific chatbots

Systems, methods, and computer program products for implementing a web crawler platform comprising one or more containerized web crawler programs working in tandem to synergistically index web resources and reduce redundancy experienced by multiple web crawlers independently indexing overlapping web resources. The platform provides a URL namespace, allowing crawlers to register with the platform and create URL endpoints for other crawlers to discover existing crawlers registered to the platform and identify web resources previously indexed. The platform provides crawler to crawler communication and exchanges of data and metadata obtained from web resources that have been previously indexed, allowing for crawlers to share existing data or metadata without having to directly crawl through the web resource. As web crawlers move between data centers of different geolocations, the crawler's registered URL is mapped to subsequent IP addresses, allowing for transparency and continuous identification by other crawlers registered with the platform.

Web crawler platform

A process for context-based autocompletion suggestion includes, based on initiating an online chat by a user of a computer system, the chat having as participants the user and a target participant, providing participant information that identifies the participants. The process maintains at the computer system a relation-based autocompletion model for use in word autocompletion suggesting. The maintained relation-based autocompletion model is initially a template model based on to relationship between chat participants, and is updated over time via training. The process suggests to the user words for autocompletion in the user's composition of chat messages, the suggested words being selected as suggestions based on the maintained relation-based autocompletion model. Maintaining the relation-based autocompletion model includes sending chat messages as training data to further train the autocompletion model.

Context-based autocompletion suggestion

A server and method for converting a textual data prompt embedded within a graphical user interface (GUI) to a widget. Parameters of a textual data prompt, which was previously received from a first computing device, are compared to additional parameters associated with widget. A first match and a second match between the parameters and the additional parameters associated with a first widget and a second widget, respectively, of the widgets is identified. In response to determining that a percentage of the first match exceeds a confidence threshold and a percentage of the second match fails to exceed the confidence threshold, the first widget embedded within the GUI on a second computing device is generated. A first response to the textual data prompt by a user associated with the second computing device is received. The first response is displayed as textual data embedded within the GUI of the first computing device.

Automatically converting a textual data prompt embedded within a graphical user interface (GUI) to a widget

A computer system including program instructions to receive a message at a source gateway of the local network, the message includes message data corresponding to a plurality of message elements, assign a unique group ID based on the type of message received at the source gateway, extract a message format from the received message, the message format defines how the message data is organized with respect to the message elements, and associate the extracted message format with the unique group ID. The computer system further including program instructions to store locally, the extracted message format together with the associated unique group ID, establish a dedicated connection between the source gateway and a target gateway of the remote network based on the unique group ID, encode the message based on the extracted message format, and send the encoded message from the source gateway to the target gateway across the dedicated connection.

Delivery of incremental sensor data over optimized channel

A recipient recommendation system and method may include receiving an input from a message creator using a messaging platform, continuously analyzing the input for a plurality of dynamic characteristics, as the input is received from the message creator via the messaging platform, the plurality of dynamic characteristics being used to determine a content of the input, comparing the content with an interest map that aggregates interests of a plurality of potential recipients from of a plurality of interest matrices, wherein the interests of the plurality of potential recipients are acquired by constantly monitoring a message platform activity of the plurality of potential recipients, determining, as a function of the comparing, a relevance ranking of the plurality of potential recipients with respect to the content input by the message creator, and providing a list of recommended recipients from the plurality of recipients, based on the relevance ranking.

Real-time recommendation of message recipients based on recipient interest level in message

A method and system for improving an automated facial recognition software system is provided. The method includes automatically detecting a face of a user via an IOT device. An image of the face is retrieved and image portions are extracted from the image and represented as a vector. The user is classified via determined facial feature attributes with respect to a plurality of user type weights stored in a cache and an initial user type of the user is determined. The vector and data indicating the initial user type are transmitted to a server and a process for inferring with respect to the initial user type, the vector, and images in a specified database associated with the initial user type, a final user type of the user s performed. An identity of the user is determined based on the inferring and the identity is transmitted to the IOT device.

Automated facial recognition detection

A method including receiving a message at a source gateway of the local network, the message includes message data corresponding to a plurality of message elements, assigning a unique group ID based on the type of message received at the source gateway, extracting a message format from the received message, the message format defines how the message data is organized with respect to the message elements, and associating the extracted message format with the unique group ID. The method further including storing, locally, the extracted message format together with the associated unique group ID, establishing a dedicated connection between the source gateway and a target gateway of the remote network based on the unique group ID, encoding the message based on the extracted message format, and sending the encoded message from the source gateway to the target gateway across the dedicated connection.

Growing community of inventors

Taipei, Taiwan

Chih-Hsiung Liu