Building a Twitter Sentiment Analysis System with Recurrent Neural Networks
This paper presents a sentiment analysis solution on tweets using Recurrent Neural Networks (RNNs). The method is can classifying tweets with an 80.74% accuracy rate, considering a binary task, after experimenting with 20 different design approaches. The solution integrates an attention mechanism ai...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/7/2266 |
| _version_ | 1797540226852192256 |
|---|---|
| author | Sergiu Cosmin Nistor Mircea Moca Darie Moldovan Delia Beatrice Oprean Răzvan Liviu Nistor |
| author_facet | Sergiu Cosmin Nistor Mircea Moca Darie Moldovan Delia Beatrice Oprean Răzvan Liviu Nistor |
| author_sort | Sergiu Cosmin Nistor |
| collection | DOAJ |
| description | This paper presents a sentiment analysis solution on tweets using Recurrent Neural Networks (RNNs). The method is can classifying tweets with an 80.74% accuracy rate, considering a binary task, after experimenting with 20 different design approaches. The solution integrates an attention mechanism aiming to enhance the network, with a two-way localization system: at memory cell level and at network level. We present an in-depth literature review for Twitter sentiment analysis and the building blocks that grounded the design decisions of our solution, employed as a core classification component within a sentiment indicator of the SynergyCrowds platform. |
| first_indexed | 2024-03-10T12:57:07Z |
| format | Article |
| id | doaj.art-76454d61109445ee8da2776b515af376 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T12:57:07Z |
| publishDate | 2021-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-76454d61109445ee8da2776b515af3762023-11-21T11:47:09ZengMDPI AGSensors1424-82202021-03-01217226610.3390/s21072266Building a Twitter Sentiment Analysis System with Recurrent Neural NetworksSergiu Cosmin Nistor0Mircea Moca1Darie Moldovan2Delia Beatrice Oprean3Răzvan Liviu Nistor4Synergy Crowds OÜ, 10141 Tallin, EstoniaSynergy Crowds OÜ, 10141 Tallin, EstoniaBusiness Information Systems Department, Interdisciplinary Centre for Data Science, Babeş-Bolyai University, 400083 Cluj-Napoca, RomaniaCoaching Consult, 400191 Cluj-Napoca, RomaniaDepartment of Management, Babeş-Bolyai University, 400591 Cluj-Napoca, RomaniaThis paper presents a sentiment analysis solution on tweets using Recurrent Neural Networks (RNNs). The method is can classifying tweets with an 80.74% accuracy rate, considering a binary task, after experimenting with 20 different design approaches. The solution integrates an attention mechanism aiming to enhance the network, with a two-way localization system: at memory cell level and at network level. We present an in-depth literature review for Twitter sentiment analysis and the building blocks that grounded the design decisions of our solution, employed as a core classification component within a sentiment indicator of the SynergyCrowds platform.https://www.mdpi.com/1424-8220/21/7/2266sentiment analysisrecurrent neural networktwitterclassificationattention mechanism |
| spellingShingle | Sergiu Cosmin Nistor Mircea Moca Darie Moldovan Delia Beatrice Oprean Răzvan Liviu Nistor Building a Twitter Sentiment Analysis System with Recurrent Neural Networks Sensors sentiment analysis recurrent neural network classification attention mechanism |
| title | Building a Twitter Sentiment Analysis System with Recurrent Neural Networks |
| title_full | Building a Twitter Sentiment Analysis System with Recurrent Neural Networks |
| title_fullStr | Building a Twitter Sentiment Analysis System with Recurrent Neural Networks |
| title_full_unstemmed | Building a Twitter Sentiment Analysis System with Recurrent Neural Networks |
| title_short | Building a Twitter Sentiment Analysis System with Recurrent Neural Networks |
| title_sort | building a twitter sentiment analysis system with recurrent neural networks |
| topic | sentiment analysis recurrent neural network classification attention mechanism |
| url | https://www.mdpi.com/1424-8220/21/7/2266 |
| work_keys_str_mv | AT sergiucosminnistor buildingatwittersentimentanalysissystemwithrecurrentneuralnetworks AT mirceamoca buildingatwittersentimentanalysissystemwithrecurrentneuralnetworks AT dariemoldovan buildingatwittersentimentanalysissystemwithrecurrentneuralnetworks AT deliabeatriceoprean buildingatwittersentimentanalysissystemwithrecurrentneuralnetworks AT razvanliviunistor buildingatwittersentimentanalysissystemwithrecurrentneuralnetworks |