公开(公告)号：US20170308790A1

公开(公告)日：2017-10-26

申请号：US15134719

申请日：2016-04-21

申请人： International Business Machines Corporation

发明人： Cicero Nogueira dos Santos ,  Bing Xiang ,  Bowen Zhou

IPC分类号： G06N3/08 ,  G06N99/00

CPC分类号： G06N3/084 ,  G06N3/0454

摘要： According to an aspect a method includes configuring a convolutional neural network (CNN) for classifying text based on word embedding features into a predefined set of classes identified by class labels. The predefined set of classes includes a class labeled none-of-the-above for text that does not fit into any of the other classes in the predefined set of classes. The CNN is trained based on a set of training data. The training includes learning parameters of class distributed vector representations (DVRs) of each of the predefined set of classes. The learning includes minimizing a pair-wise ranking loss function over the set of training data. A class embedding matrix of the class DVRs of the predefined set of classes that excludes a class embedding for the none-of-the-above class is generated. Each column in the class embedding matrix corresponds to one of the predefined classes.

公开(公告)号：US09659248B1

公开(公告)日：2017-05-23

申请号：US15000647

申请日：2016-01-19

申请人： International Business Machines Corporation

发明人： Luciano de Andrade Barbosa ,  Daria Bogdanova ,  Matthias Kormaksson ,  Cicero Nogueira dos Santos ,  Bianca Zadrozny

IPC分类号： G06N3/04 ,  G06N3/08 ,  G06F17/28

CPC分类号： G06N3/0472 ,  G06F17/2785 ,  G06F17/28 ,  G06F17/30654 ,  G06F17/30684 ,  G06N3/0427 ,  G06N3/0454 ,  G06N3/08 ,  G06N3/084

摘要： Determining semantically equivalent text or questions using hybrid representations based on neural network learning. Weighted bag-of-words and convolutional neural networks (CNN) based distributed vector representations of questions or text may be generated to compute the semantic similarity between questions or text. Weighted bag-of-words and CNN based distributed vector representations may be jointly used to compute the semantic similarity. A pair-wise ranking loss function trains neural network. In one embodiment, the parameters of the system are trained by minimizing a pair-wise ranking loss function over a training set using stochastic gradient descent (SGD).

公开(公告)号：US20210366580A1

公开(公告)日：2021-11-25

申请号：US16880021

申请日：2020-05-21

申请人： International Business Machines Corporation

发明人： Payel Das ,  Flaviu Cipcigan ,  Kahini Wadhawan ,  Inkit Padhi ,  Enara C Vijil ,  Pin-Yu Chen ,  Aleksandra Mojsilovic ,  Tom D.J. Sercu ,  Cicero Nogueira dos Santos

IPC分类号： G16C20/70 ,  G16H50/20 ,  G16H70/40 ,  G16H50/50 ,  G16H10/40 ,  G16H70/60 ,  G06F30/27

摘要： Techniques for filtering artificial intelligence (AI)-designed molecules for laboratory testing provided. According to an embodiment, computer implemented method can comprise selecting, by a system operatively coupled to a processor, a first subset of AI-designed molecules from a set of AI-designed molecules as candidate pharmaceutical agents based on classification of the AI-designed molecules using one or more classifiers. The method further comprises selecting, by the system, a second subset of the candidate pharmaceutical agents for wet laboratory testing based on evaluation of molecular interactions between the candidate pharmaceutical agents and one or more biological targets using one or more computer simulations.

公开(公告)号：US20160210310A1

公开(公告)日：2016-07-21

申请号：US14598776

申请日：2015-01-16

申请人： International Business Machines Corporation

发明人： Cicero Nogueira dos Santos ,  Marcos R. Vieira ,  Bianca Zadrozny

IPC分类号： G06F17/30

CPC分类号： G06F16/29 ,  G06F16/2477

摘要： In an approach for extracting geospatial temporal facts and events, a processor receives a set of structured data and a set of unstructured data. A processor extracts a first set of temporal information and a first set of geospatial information from the set of unstructured data. A processor identifies a second set of temporal information and a second set of geospatial information from the set of structured data. A processor determines that the set of structured data and the set of unstructured data are related, based on at least the first set of temporal information, the second set of temporal information, the first set of geospatial information, and the second set of geospatial information. A processor groups the set of structured data and the set of unstructured data into a collective set of data. A processor stores the collective set of data.

摘要翻译： 在提取地理空间时间事件和事件的方法中，处理器接收一组结构化数据和一组非结构化数据。 处理器从非结构化数据集中提取第一组时间信息和第一组地理空间信息。 处理器从所述一组结构化数据识别第二组时间信息和第二组地理空间信息。 处理器基于至少第一组时间信息，第二组时间信息，第一组地理空间信息和第二组地理空间信息来确定结构化数据集和非结构化数据集是相关的 。 A处理器将该组结构化数据和一组非结构化数据分组为一组集合的数据。 处理器存储集合的数据。