International Business Machines Corporation

Junchi Yan

Min Gong

Ning Duan

Yu Ying Wang

Chao Zhang

Yu Wang

Stephen Mingyu Chu

Wei Shan Dong

Xin Zhang

Chunhua Tian

Chen Wang

Xuan Zhang

Liangliang Cao

Yun Jie Qiu

Yaoliang Chen

Xiaomin Xu

CaiQin Ma

Qian Lin

Dong Sheng Li

Chang Li

Guoqiang Hu

Jun Zhu

Jian Ding

Zhi Hu Wang

Peng Gao

Juhnyoung Lee

Yi Qi

Ya Ming Zhang

Jing Chang Huang

Chao Xue

Jin Huang

Qian Zhao

Jian Dong Ding

Weipeng Zhang

Boliang Chen

Fan Yang

Wei Zhao

Chun Ma

Yue Hua Zhang

Yao Liang Chen

Hui Zhu

Sheng Huang

Wen Ji Huang

Yuan Ni

Wen Ting Mo

Ren Jie Yao

Wen Peng Xiao

Baohua Yang

Wei Peng Zhang

Xiu Fang Zhu

A method and apparatus for generating a training model based on feedback are provided. The method for generating a training model based on feedback, includes calculating an eigenvector of a sample among a plurality of samples; obtaining scores granted by a user for one or more of the plurality of samples in a round, obtaining scores granted by the user for a first number of samples; obtaining scores granted by the user for a second number of samples in response to detecting, based on the eigenvector, an inconsistency between the scores granted by the user for the first number of samples; and generating a training model based on the scores granted by the user for the first and second numbers of samples. A corresponding apparatus is also provided.

Generating a training model based on feedback

Techniques for neural network training are provided. One computer-implemented method comprises: obtaining, by an electronic device operatively coupled to a processing unit, based on a similarity between two images, a first image and a second image. The computer-implemented method also comprises training, by the electronic device, a neural network based on the first image and the second image such that a distance between a first vector and a second vector generated respectively from the first image and the second image in the trained neural network is associated with the similarity.

Neural network training

Techniques for model based data processing are proposed. In one example, a computer-implemented method is as follows. A first model is determined from a first category of data processing models and a second model is determined from a second category of data processing models. The second category can be different from the first category in some instances. Performance of a first combined model of the first and second models can be compared with performance of the first model. A target model for processing data can be selected from the first combined model and the first model based on the comparing.

Model based data processing

One or more sensors associated with a vehicle detect a roadway condition of a first roadway and an evasive maneuver is performed by the vehicle to avoid the detected roadway condition. In response to one or more processors determining that the evasive maneuver was successful, a record of the successful maneuver and the roadway condition are stored in a database. Subsequent to storing the record in the database, one or more computers associated with one or more vehicles is trained to execute the evasive maneuver, in response to determining that the one or more vehicles is exposed to the roadway condition experienced by the vehicle.

Training a self-driving vehicle

A method and apparatus for generating a training model based on feedback are provided. The method for generating a training model based an feedback, includes calculating an eigenvector of a sample among a plurality of samples; obtaining scores granted by a user for one or more of the plurality of samples in a round, obtaining scores granted by the user for a first number of samples; obtaining scores granted by the user for a second number of samples in response to detecting, based on the eigenvector, an inconsistency between the scores granted by the user for the first number of samples; and generating a training model based on the scores granted by the user for the first and second numbers of samples. A corresponding apparatus is also provided.

A method, system and computer program product are disclosed for recommending terms in a document based on a specified interaction with the document. In one embodiment, the method comprises displaying a document on a device; detecting a specified interaction with the document displayed on the device; identifying text associated with the specified interaction with the document displayed on the device; performing object recognition with the identified text to recognize one or more defined interested objects; and recommending terms in the document based on the recognized defined interested objects. In an embodiment, the object recognition includes identifying one or more objects from the text, and performing the object recognition with these objects to recognize the one or more interested objects. In an embodiment, one or more terms are extracted from the text, and these terms are partitioned to identify the one or more objects from the text.

Interest highlight and recommendation based on interaction in long text reading

The present invention relates to a method and apparatus for genome compression and decompression. In one embodiment of the present invention, there is a method for genome compression, including: selecting from a reference database a reference genome that matches the genome; building an index based on positions of the reference genome's multiple segments in the reference genome; aligning the genome with the reference genome based on the multiple segments so as to identify difference data between the genome and the reference genome; and generating a compressed genome, the compressed genome including at least the index and the difference data. In other embodiments, there is provided an apparatus for genome compression. Further, there is a method and apparatus for decompressing the genome that has been compressed using the above method and apparatus.

Genome compression and decompression

A method of generating a predictor to classify data includes: training each of a plurality of first classifiers arranged in a first level on current training data; operating each classifier of the first level on the training data to generate a plurality of predictions; combining the current training data with the predictions to generated new training data; and training each of a plurality of second classifiers arranged in a second level on the new training data. The first classifiers are classifiers of different classifier types, respectively and the second classifiers are classifiers of the different classifier types, respectively.

Multi-layer information fusing for prediction

An approach for a computer to evaluate a predictive model includes identifying features of training samples in a set of training samples. The approach selects evaluation metrics from a set of evaluation metrics as available metrics using identified features and includes determining recommended metrics using the predictive model, the available metrics, and a predetermined set of user-preferred metrics. The approach applies the predictive model created using the set of training samples to a set of test samples to calculate values of the available metrics. The approach evaluates the predictive model by using the available metrics and the values of the available metrics to evaluate the predictive model by evaluating the predictive model using the recommended metrics and the values of the recommended metrics.

Method and apparatus for evaluating predictive model

Processing GPS drifting. A method and system for processing GPS drifting is included. The method includes: determining predicted error information on position correction of a device within a region; obtaining a GPS initial position and a corrected position of the device within the region; and determining GPS drifting of the region based on the predicted error information, the GPS initial position, and the corrected position of at least one device within the region. The method and system of the present application can effectively ameliorate GPS drifting issues.

Processing GPS drifting

A method and apparatus for determining similarity among gene sequences, for compressing a gene sequence, and for decompressing a gene sequence. The method for determining similarity between a first gene sequence and a second gene sequence includes: moving a sliding window of a predefined length on the first gene sequence and the second gene sequence respectively; extracting a first part Stringof the first gene sequence within the sliding window, and a second part Stringof the second gene sequence within the sliding window during the imovement of the sliding window; and determining similarity between the first gene sequence and the second gene sequence based on the first part Stringand the second part String. Also provided is an apparatus for the above method.

Managing gene sequences

Systems for database optimization include a database rebuild module configured to compact columns in an unoptimized database that co-appear in queries to the unoptimized database into a single hyper-column to form an optimized database. A database interface module is configured to accept queries according to a schema for the unoptimized database and to translate the accepted queries into queries according to a schema for the optimized database.

Optimizing tables with too many columns in a database

Methods and systems for database optimization include compacting columns in an unoptimized database that co-appear in queries to the database into a single hyper-column using a processor to form an optimized database. Queries are received according to a schema for the unoptimized database. The received queries are translated into queries according to a schema for the optimized database.

A method for determining an object referenced within a set of one or more informal online communications comprises: generating a knowledge graph for a company based at least on formal online communications, the knowledge graph comprising a plurality of node elements, and the knowledge graph further comprising, for each node element of the knowledge graph, a corresponding halo comprising one or more words which are temporally proximate to that node element within the formal online communications; for each node element of the knowledge graph which is determined to be present in a given informal online communication, detecting a halo comprising one or more words which are temporally proximate to that node element within the given informal online communication; and identifying which of the plurality of node elements has a corresponding halo within the knowledge graph most similar to the detected halo, wherein the identified node element is the referenced object.

Determining an object referenced within informal online communications

Growing community of inventors

Shanghai, China

Junchi Yan