International Business Machines Corporation

Yan Feng Han

Qi Ruan

Kun Yan Yin

Yan Bin Hu

Jian Hua Lv

Lei Mei

Peng Hui Jiang

Wen Jie Wang

Wei Yan

Yan Nan Hu

Jian Yu Wang

Lian Na Wang

Bei Bei Zhan

Yun Ze Miao

Kai Feng Wang

A computer-implemented method includes receiving a viewer request for playing a video summary of a video, wherein the viewer request includes a length of the video summary, generating the video summary of the viewer-requested length comprising a set of frames selected from the video based on audience reviews of the video, and playing a video stream of the video summary.

Generating video summary

A user input in a source language is received. A set of contextual data is received. The user input is encoded into a user input feature vector. The set of contextual data is encoded into a context feature vector. The user input feature vector and the context feature vector are used to generate a fusion vector. An adaptive neural network is trained to identify a second context feature vector, based on the fusion vector. A second user input in the source language is received for translation into a target language. The adaptive neural network is used to determine, based on the second context feature vector, a second user input feature vector. The second user input feature vector is decoded, based on the source language and the target language, into a target language output. A user is notified of the target language output.

Adaptive language translation using context features

Methods, computer program products, and systems for content rendering are provided. An exemplary method comprises obtaining a current scrolling speed; determining a current speed level based on the current scrolling speed; determining a current rendering mode based on the determined current speed level; and rendering at least one component based on the current rendering mode.

Speed-based content rendering

A computer-implemented method comprises analyzing content sections on each of a plurality of open browser pages using natural language processing to identify one or more topics on each of the plurality of open browser pages; calculating a respective relevance score for each of the content sections; grouping each of the plurality of topics into one of a plurality of topic groups; calculating a respective group ranking for each of the plurality of topic groups based on the respective relevance score for each content section. The method further comprises, for each topic group, assigning the respective group ranking to all of the content sections corresponding to the respective topic group; and, for each of the plurality of open browser pages, selecting at least one content section having a highest group ranking and modifying a display of the respective open browser page to direct attention to the selected content section.

Page content ranking and display

Aspects of the present invention disclose a method, computer program product, and system for content rendering. The method includes one or more processors retrieving at least one user interface (UI) component from a Document Object Model (DOM) tree. The method further includes one or more processors determining a corresponding rendering priority level (RPL) for each of the at least one UI component. The method further includes, in response to determining that the corresponding RPL of a first UI component of the at least one UI component is above a first threshold, one or more processors rendering the first UI component.

Priority-based rendering

A method and an apparatus for classifying spoken content in a teleconference for a follower of the teleconference is disclosed. The method comprises: detecting a topic to which the spoken content belongs; determining a (overall) correlation degree between the follower and the spoken content at least according to a correlation degree between the follower and the topic; and classifying the spoken content according to the (overall) correlation degree between the follower and the spoken content. With the method and the apparatus, the correlation degree between the spoken content in the teleconference and the follower of the teleconference can be determined automatically, and the spoken content can be classified according to the correlation degree, so that the follower can selectively pay attention to some spoken contents during the teleconference, which reduces a burden of the follower and improves conference efficiency.

Growing community of inventors

Ningbo, China

Yan Feng Han