Filters








30,283 Hits in 6.4 sec

A Comparison of Approaches for Measuring the Semantic Similarity of Short Texts Based on Word Embeddings

Karlo Babić, Francesco Guerra, Sanda Martinčić-Ipšić, Ana Meštrović
2020 Journal of Information and Organizational Sciences  
Since these models provide word vectors, we experiment with various methods that calculate the semantic similarity of short texts based on word vectors.  ...  In this paper, we describe a set of experiments we carried out to evaluate and compare the performance of different approaches for measuring the semantic similarity of short texts.  ...  Kenter and De Rijke proposed measuring the semantic similarity of short texts by combining word embeddings with external knowledge sources [11] .  ... 
doi:10.31341/jios.44.2.2 fatcat:7hmceah4q5bqjl75mumzgyls4e

Cognitive Aspects-Based Short Text Representation with Named Entity, Concept and Knowledge

Wenfeng Hou, Qing Liu, Longbing Cao
2020 Applied Sciences  
ECKA is a multi-level short text semantic representation model, which extracts the semantic features from the word, entity, concept and knowledge levels by CNN, respectively.  ...  One important solution is to enrich short text representation by involving cognitive aspects of text, including semantic concept, knowledge, and category.  ...  Finally, as the word, entity and concept embeddings with 300 dimensions are learned by word2vec and the knowledge graph embedding with 50 dimensions is learned from TransE, the two embeddings need to be  ... 
doi:10.3390/app10144893 fatcat:uotbuk4tajcghmmwrlnkbkrfze

Psychiatric symptom recognition without labeled data using distributional representations of phrases and on-line knowledge

Yaoyun Zhang, Olivia Zhang, Yonghui Wu, Hee-Jin Lee, Jun Xu, Hua Xu, Kirk Roberts
2017 Journal of Biomedical Informatics  
Finally, semantic similarity between the distributional representations of the seed symptoms and candidate symptoms is calculated to assess the relevance of a phrase.  ...  Results & Conclusion-Our method demonstrates good performance at extracting symptoms from an unseen corpus, including symptoms with no word overlap with the provided seed terms.  ...  , knowledge-based methods using external sources of structured semantic knowledge, and corpus-based methods such as distributional semantics.  ... 
doi:10.1016/j.jbi.2017.06.014 pmid:28624644 pmcid:PMC5705397 fatcat:vmqd6edvxfhgfmrfzvy4k36cla

Topic Modeling for Short Texts via Word Embedding and Document Correlation

Feng Yi, Bo Jiang, Jianjun Wu
2020 IEEE Access  
In fact, each short text usually contains a limited number of topics, and understanding semantic content of short text needs to the relevant background knowledge.  ...  TRNMF integrates successfully both word co-occurrence regularization and sentence similarity regularization into topic modeling for short texts.  ...  [39] propose a global and local word embedding-based topic model (GLTM) for short texts, where the global word embeddings is learned from large external corpus and the local word embeddings is obtained  ... 
doi:10.1109/access.2020.2973207 fatcat:qrmkhfoxqjb4bcutcyicuwvpiy

A Paraphrase Identification Approach in Paragraph length texts

Arwa Al Saqaabi, Craig Stewart, Eleni Akrida, Alex, Ra Cristea, Antonija Mitrovic, Nigel Bosch
2022 Zenodo  
How to measure the semantic similarity of natural language is a fundamental issue in many tasks, such as paraphrase identification (PI) and plagiarism detection (PD) which are intended to solve major issues  ...  There are many approaches that have been suggested, such as machine learning (ML) and deep learning (DL) methods.  ...  They also compare similarity metrics with two types of word embedding deep learning models: word2vec and Fast-Text [27] .  ... 
doi:10.5281/zenodo.6852990 fatcat:xhm6lzqkszb4xg6xadqp22heha

A Tri-Partite Neural Document Language Model for Semantic Information Retrieval [chapter]

Gia-Hung Nguyen, Lynda Tamine, Laure Soulier, Nathalie Souf
2018 Lecture Notes in Computer Science  
Previous work in information retrieval have shown that using evidence, such as concepts and relations, from external knowledge sources could enhance the retrieval performance.  ...  This paper presents a new tri-partite neural document language framework that leverages explicit knowledge to jointly constrain word, concept, and document learning representations to tackle a number of  ...  -Knowledge source-based context view (A2): constraining the learning of word-concept pairs with respect to a knowledge source structure allows obtaining close word embeddings for words sharing the same  ... 
doi:10.1007/978-3-319-93417-4_29 fatcat:rxoma64ekrc5xhgzvgwttxj4zy

Toward a Deep Neural Approach for Knowledge-Based IR [article]

Gia-Hung Nguyen, Lynda Tamine, Laure Soulier, Nathalie Bricon-Souf
2016 arXiv   pre-print
This latter issue is tackled by recent works dealing with deep representation learn ing of texts.  ...  With this in mind, we argue that embedding KBs within deep neural architectures supporting documentquery matching would give rise to fine-grained latent representations of both words and their semantic  ...  In the same mind, Severyn and Moschitti [16] present another convolutional neural network architecture to learn the optimal representation of short text pairs as well as the similarity function.  ... 
arXiv:1606.07211v1 fatcat:jdypcyno3zcwphnoclk44dsfxi

Recommendation Model Based on Semantic Features and a Knowledge Graph

Yudong Liu, Wen Chen, Zhihan Lv
2021 Wireless Communications and Mobile Computing  
More specifically, we first select DBpedia as a knowledge graph to extend short text features of items and get the semantic features of the items based on the extended text.  ...  And then, we calculate the item vector and further obtain the semantic similarity degrees of the users.  ...  This work was supported by the project of the Guangzhou Science and Technology Bureau, China, under Grant no. 202007040006. 8 Wireless Communications and Mobile Computing  ... 
doi:10.1155/2021/2382892 fatcat:ii2iuzahafdphlfe2f2fk7ayba

Semantic enriched deep learning for document classification

Abderrahmane Larbi, Sarra Ben Abbès, Rim Hantach, Lynda Temal, Philippe Calvez
2020 Joint Ontology Workshops  
The experiments demonstrate that the proposed hybrid architecture with the additional semantic knowledge improves the results.  ...  This approach was compared to some state-of-the-art text classification approaches not including semantic knowledge.  ...  Bian et al [20] present a leverage morphological, syntactic, and semantic knowledge to achieve high-quality word embeddings and prove that knowledge-powered deep learning can enhance their effectiveness  ... 
dblp:conf/jowo/LarbiAHTC20 fatcat:3yaq454bnbe3tc7iwddod3srx4

Multi-view and Multi-source Transfers in Neural Topic Modeling with Pretrained Topic and Word Embeddings [article]

Pankaj Gupta, Yatin Chaudhary, Hinrich Schütze
2019 arXiv   pre-print
Though word embeddings and topics are complementary representations, several past works have only used pre-trained word embeddings in (neural) topic modeling to address data sparsity problem in short text  ...  Then, we identify one or multiple relevant source domain(s) and take advantage of corresponding topics and word features via the respective pools to guide meaningful learning in the sparse target domain  ...  semantics) exclusively or jointly from one or many sources (i.e., multi-view and multi-source) that better deal with data-sparsity issues, especially in a short-text and/or small document collection.  ... 
arXiv:1909.06563v2 fatcat:42szwe4psza25fgmwnhlqfi32q

Leveraging Knowledge-Based Features With Multilevel Attention Mechanisms for Short Arabic Text Classification

Iyad Alagha
2022 IEEE Access  
A common solution is to enrich the short text with additional semantic features extracted from external knowledge, such as Wikipedia, to help the classifier better decide on the correct class.  ...  A deep learning model with multiple attention mechanisms is then used to encode the short text and the associated category set.  ...  They examined different approaches to text augmentation, such as augmenting the text with synonyms or semantically similar words at either the source text or the embedding levels.  ... 
doi:10.1109/access.2022.3175306 fatcat:jwn327xofzdyjpdcmrgo6mjyhy

A Hybrid Classification Method via Character Embedding in Chinese Short Text with Few Words

Yi Zhu, Yun Li, Yongzheng Yue, Jipeng Qiang, Yunhao Yuan
2020 IEEE Access  
More specifically, firstly, the character embedding is computed to represent Chinese short texts with few words, which takes full advantage of short text information without external corpus.  ...  Meanwhile, contemporary short text classification methods either to expand feature of short text with external corpus or to learn the feature representation from all the texts, which have not take the  ...  Due to that not all words have a positive effect on classification especially in the Chinese short texts with few words, the reservation of notional words and semantic similarity calculation are adopted  ... 
doi:10.1109/access.2020.2994450 fatcat:vfcj2ocm55fidp46vsxs4p6ucu

Understanding Short Texts [chapter]

Haixun Wang
2013 Lecture Notes in Computer Science  
semantics (embedding and deep learning).  ...  deep learning) to model latent semantic representation for short texts.  ... 
doi:10.1007/978-3-642-37401-2_1 fatcat:kmy4l3pudzbujfn75a45msmgqa

Mining and searching association relation of scientific papers based on deep learning [article]

Jie Song and Meiyu Liang and Zhe Xue and Feifei Kou and Ang Li
2022 arXiv   pre-print
Therefore, the research on mining and searching the association relationship of scientific papers based on deep learning has far-reaching practical significance.  ...  The phenomenon reveals the data characteristics, laws, and correlations contained in the data of scientific and technological papers in specific fields, which can realize the analysis of scientific and  ...  Compared with traditional embedding models, deep semantic representation learning [53] combines entity-rich side information (e.g. multimodal information, knowledge graph, meta information, etc.) with  ... 
arXiv:2204.11488v1 fatcat:zxwvpnids5bopberzumjofgupq

Deep Short Text Classification with Knowledge Powered Attention [article]

Jindong Chen, Yizhou Hu, Jingping Liu, Yanghua Xiao, Haiyun Jiang
2019 arXiv   pre-print
In this paper, we retrieve knowledge from external knowledge source to enhance the semantic representation of short texts.  ...  And we classify a short text with the help of conceptual information.  ...  This allows the model to retrieve knowledge from an external knowledge source that is not explicitly stated in the short text but relevant for classification.  ... 
arXiv:1902.08050v1 fatcat:pvwybep5rzc3tgw2wb47r5a2le
« Previous Showing results 1 — 15 out of 30,283 results