Patent
Word vector processing methods, word vector processing device and electronic equipment
Zhou Jun
- 12 Jan 2018
2
TL;DR: In this paper, the embodiment of the description discloses word vector processing methods, a word vector processor and electronic equipment, including word vectors of words on the basis of character vectors of the n-ary characters for training, wherein the nary character characterizes n consecutive characters of the corresponding word thereof.
read more
Abstract: The embodiment of the description discloses word vector processing methods, a word vector processing device and electronic equipment. One of the methods includes: obtaining a plurality of n-ary characters from words by dividing, and using word vectors of the words on the basis of character vectors of the n-ary characters for training, wherein the n-ary character characterizes n consecutive characters of the corresponding word thereof, and for Japanese words taken as an example, the characters thereof are kana and/or kanji.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Word vector generation method, device and equipment
Cao Shaosheng,Zhou Jun +1 more
- 13 Aug 2019
TL;DR: In this article, a word vector generation method, device and equipment is described. The method comprises the steps of obtaining words obtained by word segmentation of a corpus, establishing feature vectors of the words based on n-element characters, training a convolutional neural network according to the feature vectors and the feature vector of context words of words in the corpus, and generating word vectors of words according to a trained CNN.
1
Patent
Word embedding method and device based on double-layer attention mechanism and storage device
Hong Yao,Chen Renyao,Liu Chao,Lijun Dong,Xiaojun Kang,Li Xinchuan,Li Shengwen,Liang Qingzhong,Zheng Kun +8 more
- 20 Aug 2019
TL;DR: In this paper, a double-layer attention mechanism and a storage device are used for word embedding in vocabularies, which can capture semantic weights inside the vocabulary and semantic original weights inside semantics, thus, the change of the semantics inside the context can be described more deeply and accurately.
References
Patent
Word vector training method and device
Xing Ning,Liu Mingrong,Xu Jingfang,Chang Xiaofu,Wang Xiaowei +4 more
- 20 Jul 2016
TL;DR: In this article, a distributed word vector learning model is configured to perform periodic word vector training on each word in the word list, and the word vector corresponding to the word in a word list is obtained.
23
Patent
Method, device and server for obtaining similarity of key words
Tang Huang
- 08 Oct 2014
TL;DR: In this article, a method, device and server for obtaining the similarity of key words, and belongs to the field of information technology, is described, which comprises the steps: obtaining key words of user labels and key words in the interested classes; computing a distance between the word vector of each key word in the key words and the word vectors of each word in each keyword in the user labels according to the key word of the user label and the words vector of the keyword of each keyword in the interest classes; and obtaining the distance between a first key word and the
21
Patent
Learning entity and word embeddings for entity disambiguation
Chen Zheng,Zhang Jianwen +1 more
- 02 May 2018
TL;DR: The authors proposed a probabilistic model for the one or more concurrence graphs, defining an objective function based on the model and the concurrence graph, and training at least one disambiguation model using feature vectors generated through an optimized version of the objective function.
5
Patent
Full-text retrieval device and method for interactive electronic technical manual of shipping equipment
Ma Liangli,Tan Jiwei,Su Kai,Xu Guopeng +3 more
- 27 Apr 2016
TL;DR: In this paper, a full-text retrieval device for an interactive electronic technical manual of shipping equipment is presented, which comprises a common source database, a specialized vocabulary extraction module, an abbreviation extraction module and a first segmentation module.
2