7,503 Hits in 5.1 sec

Leveraging Aspect Phrase Embeddings for Cross-Domain Review Rating Prediction [article]

Aiqi Jiang, Arkaitz Zubiaga
2018 arXiv   pre-print
We introduce a model that leverages aspect phrase embeddings extracted from the reviews, which enables the development of both in-domain and cross-domain review rating prediction systems.  ...  The cross-domain review rating prediction system is particularly significant for the least popular review domains, where leveraging training data from other domains leads to remarkable improvements in  ...  We propose to pursue the review rating prediction for non-popular domains by developing a cross-domain rating prediction system for the first time, where rated reviews from popular domains can be leveraged  ... 
arXiv:1811.05689v1 fatcat:7742lznberesza4t2w33fesupm

Leveraging aspect phrase embeddings for cross-domain review rating prediction

Aiqi Jiang, Arkaitz Zubiaga
2019 PeerJ Computer Science  
We introduce a model that leverages aspect phrase embeddings extracted from the reviews, which enables the development of both in-domain and cross-domain review rating prediction systems.  ...  The cross-domain review rating prediction system is particularly significant for the least popular review domains, where leveraging training data from other domains leads to remarkable improvements in  ...  CONCLUSIONS In this work we have proposed a novel method that leverages aspect phrase embeddings for predicting the star rating of online reviews, which is applicable in in-domain and cross-domain settings  ... 
doi:10.7717/peerj-cs.225 pmid:33816878 pmcid:PMC7924723 fatcat:hu4ouacsznarhlqxl2pciqy2w4

Domain-oriented Language Pre-training with Adaptive Hybrid Masking and Optimal Transport Alignment [article]

Denghui Zhang, Zixuan Yuan, Yanchi Liu, Hao Liu, Fuzhen Zhuang, Hui Xiong, Haifeng Chen
2021 arXiv   pre-print
To address the above issues, we provide a generalized domain-oriented approach, which leverages auxiliary domain knowledge to improve the existing pre-training framework from two aspects.  ...  Second, we introduce Cross Entity Alignment to leverage entity association as weak supervision to augment the semantic learning of pre-trained models.  ...  Following [36] , we apply a dense layer and softmax layer on top of BERT output embeddings to predict the sequence labels. Review Aspect Sentiment Classification (Review ASC).  ... 
arXiv:2112.03024v1 fatcat:g4xqxd5lrzctfd4eikp56jakoe

E-BERT: A Phrase and Product Knowledge Enhanced Language Model for E-commerce [article]

Denghui Zhang, Zixuan Yuan, Yanchi Liu, Fuzhen Zhuang, Haifeng Chen, Hui Xiong
2021 arXiv   pre-print
., review-based question answering, aspect extraction, aspect sentiment classification, and product classification.  ...  To utilize product-level knowledge, we introduce Neighbor Product Reconstruction, which trains E-BERT to predict a product's associated neighbors with a denoising cross attention layer.  ...  ., review-based question various domain phrases.  ... 
arXiv:2009.02835v3 fatcat:w4p4ldnwd5g3tomjiqdw5ndkla

A Knowledge-Driven Approach to Classifying Object and Attribute Coreferences in Opinion Mining [article]

Jiahua Chen and Shuai Wang and Sahisnu Mazumder and Bing Liu
2021 arXiv   pre-print
., product aspects) in opinionated reviews is crucial for improving the opinion mining performance.  ...  This paper proposes an approach to automatically mine and leverage domain-specific knowledge for classifying objects and attribute coreferences.  ...  To make the knowledge base fit for deep neural models, we concatenate SenticNet embeddings with BERT embeddings to extend the embedding information. 3. Domain-specific KB.  ... 
arXiv:2010.05357v2 fatcat:usc6gsgvunbcdlo4me52ixqudm

Fine-Grained Opinion Summarization with Minimal Supervision [article]

Suyu Ge, Jiaxin Huang, Yu Meng, Sharon Wang, Jiawei Han
2021 arXiv   pre-print
An opinion-oriented spherical word embedding space is trained to provide weak supervision for the phrase classifier, and phrase clustering is performed using the aspect-aware contextualized embedding generated  ...  Each cluster consists of semantically coherent phrases, expressing uniform opinions towards certain sub-aspect or characteristics (e.g., positive feelings for "burgers" in the "food" aspect).  ...  Given the aspect/sentiment set and keyword lists, we leverage them as seeds to learn text embeddings tailored for aspects/sentiments.  ... 
arXiv:2110.08845v1 fatcat:eoeagv4z6zfexa6jiuc5cektle

Constructing Explainable Opinion Graphs from Review [article]

Nofar Carmeli and Xiaolan Wang and Yoshihiko Suhara and Stefanos Angelidis and Yuliang Li and Jinfeng Li and Wang-Chiew Tan
2021 arXiv   pre-print
search over opinion phrases.  ...  We present ExplainIt, a system that extracts and organizes opinions into an opinion graph, which are useful for downstream applications such as generating explainable review summaries and facilitating  ...  rate.  ... 
arXiv:2006.00119v2 fatcat:lqlsxieoivfnxad7i65skfpia4

CATN: Cross-Domain Recommendation for Cold-Start Users via Aspect Transfer Network [article]

Cheng Zhao, Chenliang Li, Rong Xiao, Hongbo Deng, Aixin Sun
2020 arXiv   pre-print
To this end, we propose a cross-domain recommendation framework via aspect transfer network for cold-start users (named CATN).  ...  CATN is devised to extract multiple aspects for each user and each item from their review documents, and learn aspect correlations across domains with an attention mechanism.  ...  Then we can highlight the important aspect pairs at domain level for better rating prediction. To this end, we design a simple but effective method for cross-domain preference matching.  ... 
arXiv:2005.10549v1 fatcat:pmgzfsaeozgp7cjjzoz2exh7gy

Analyzing Sentiments in One Go: A Supervised Joint Topic Modeling Approach

Zhen Hai, Gao Cong, Kuiyu Chang, Peng Cheng, Chunyan Miao
2017 IEEE Transactions on Knowledge and Data Engineering  
In this work, we focus on modeling user-generated review and overall rating pairs, and aim to identify semantic aspects and aspect-level sentiments from review data as well as to predict overall sentiments  ...  It also leverages sentimental overall ratings, which often comes with online reviews, as supervision data, and can infer the semantic aspects and aspect-level sentiments that are not only meaningful but  ...  Jakob and Gurevych [33] also used the CRFs model for single-domain and cross-domain feature extraction problem.  ... 
doi:10.1109/tkde.2017.2669027 fatcat:24uviz42ffhutc5ayjuvsudo4m

SEML: A Semi-supervised Multi-task Learning Framework for Aspect-Based Sentiment Analysis

Ning Li, Chi-Yin Chow, Jia-Dong Zhang
2020 IEEE Access  
embeddings and domain-specific embeddings learned from unlabeled reviews. • EMOVA [19] uses the CVT and moving-window attention mechanism to leverage both labeled and unlabeled reviews.  ...  However, for domain-dependent aspects, manual labeling could be a huge investment. One solution is to apply effective semisupervised learning to leverage a plenty of unlabeled reviews.  ... 
doi:10.1109/access.2020.3031665 fatcat:fg7taw4hfze2hnkhuewp57q2li

Zero-Shot Cross-Lingual Opinion Target Extraction [article]

Soufian Jebbara, Philipp Cimiano
2019 arXiv   pre-print
We leverage multilingual word embeddings that share a common vector space across various languages and incorporate these into a convolutional neural network architecture for OTE extraction.  ...  In this work, we address the lack of available annotated data for specific languages by proposing a zero-shot cross-lingual approach for the extraction of opinion target expressions.  ...  The approach is evaluated for English to Chinese reviews.  ... 
arXiv:1904.09122v1 fatcat:pabq54kjs5auxm63dfd2dkc26y

Toward Tag-free Aspect Based Sentiment Analysis: A Multiple Attention Network Approach [article]

Yao Qiang, Xin Li, Dongxiao Zhu
2020 arXiv   pre-print
With the Self- and Position-Aware attention mechanism, MAN is capable of extracting both aspect level and overall sentiments from the text reviews using the aspect level and overall customer ratings, and  ...  Existing aspect based sentiment analysis (ABSA) approaches leverage various neural network models to extract the aspect sentiments via learning aspect-specific feature representations.  ...  Aspect based sentiment analysis (ABSA) emerged as a more informative analysis to predict polarities based on the predefined aspect categories or aspect terms (tagged words/phrases) in the user reviews.  ... 
arXiv:2003.09986v1 fatcat:4nccxk7fxjfe3nm34c5uousnfy

EMOVA: A Semi-supervised End-to-End Moving-Window Attentive Framework for Aspect Mining [chapter]

Ning Li, Chi-Yin Chow, Jia-Dong Zhang
2020 Lecture Notes in Computer Science  
reviews from the same domain in a unified end-to-end architecture. (3) Since past nearby information in a text provides important semantic contexts for a prediction task in aspect mining, a moving-window  ...  Experimental results over four review datasets from the SemEval workshops show that EMOVA outperforms the state-of-the-art models for aspect mining.  ...  However, for domain (or even entity) dependent aspects, manual annotation could be a huge investment. One solution is to apply effective semi-supervised learning to leverage unlabeled reviews.  ... 
doi:10.1007/978-3-030-47436-2_61 fatcat:cppi7m6mljcvfbdukv6m5n4upm

Extracting Aspect Specific Opinion Expressions

Abhishek Laddha, Arjun Mukherjee
2016 Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing  
discriminative sequence modeling component for opinion phrase extraction.  ...  However, in opinion mining, it is often desirable to mine the aspect specific opinion expressions (or aspectsentiment phrases) containing both the aspect and the opinion.  ...  They can also be applied to the various tasks such as sentiment classification, comparative aspect evaluations, aspect rating prediction, etc.  ... 
doi:10.18653/v1/d16-1060 dblp:conf/emnlp/LaddhaM16 fatcat:hbfzta6kindvtb7co46so65are

Aspect and Opinion Terms Extraction Using Double Embeddings and Attention Mechanism for Indonesian Hotel Reviews [article]

Jordhy Fernando, Masayu Leylia Khodra, Ali Akbar Septiandri
2019 arXiv   pre-print
In order to extract aspect and opinion terms for Indonesian hotel reviews, we adapt double embeddings feature and attention mechanism that outperform the best system at SemEval 2015 and 2016.  ...  Using 1000 reviews for evaluation, we achieved F1-measure of 0.914 and 0.90 for aspect and opinion terms extraction in token and entity (term) level respectively.  ...  ACKNOWLEDGMENT We would like to thank AiryRooms for providing Indonesian hotel reviews used in this research to build the aspect and opinion terms extraction model.  ... 
arXiv:1908.04899v2 fatcat:otzm32zlq5a5xfecthaorddtju
« Previous Showing results 1 — 15 out of 7,503 results