![REFERENCES
• [1] A SVM-Based Method for Sentiment Analysis in Persian Language
(Mohammad Sadegh Hajmohammadi, Roliana Ibrahim)
• [2] https://gate.ac.uk/
• [3] “Majid Sezavar” http://sazvar.student.um.ac.ir/
• [4]
http://www.coreservlets.com/Apache-Tomcat-Tutorial/tomcat-7-with
-eclipse.html
• [5] http://javapapers.com/web-service/java-web-service-using-eclipse
14](https://image.slidesharecdn.com/lab-190226164805/85/Lexicon-based-Sentiment-Analysis-for-Persian-Text-14-320.jpg)
![[ASA] Sentiment Analysis in Twitter, a Study on the Saudi Community](https://cdn.slidesharecdn.com/ss_thumbnails/asa-drnoratalkslideshares4-161213191556-thumbnail.jpg?width=560&fit=bounds)
Lexicon-based Sentiment Analysis for Persian Text
- 1. LEXICON-BASED SENTIMENT ANALYSIS FOR PERSIAN TEXT Authors: Mohammadhassan Khodashahi Twitter: @KhodashahiMh Email: [email protected] 1
- 2. CONTENTS ➢ Motivation ➢ Introduction ➢ Approach ✓ Data collection ✓ Processing resource ▪ Tokenizer ▪ Sentence splitter ▪ Pos tagger ▪ Jape rules ➢ Testing ✓ Performance testing (1) ✓ Performance testing (2) ➢ Conclusion and future works ➢ References 2
- 3. MOTIVATION • Online shopping become more preferable for customers. • E-business has improved in different aspects. • User feedbacks and comments on different products have an important roll on other users’ decision. • Sentiment analysis works on this amount of data to give the expected information for different Enterprises to help them improved. 3
- 4. INTRODUCTION • Definition (Wikipedia): Sentiment analysis ( opinion mining) refers to the use of natural language processing and text mining to identify and extract subjective information in source materials. • sentiment analysis determine the attitude of a speaker as • Why for Persian language?! Number of Persian websites and users increasing, but no such research for Persian language. 4
- 5. APPROACH • Data collection: ● Gathering 7179 Persian adjectives. ● we received a total of 8278 votes for adjectives. ● Asking Persian people to vote about their sense on each adjectives ● Calculate the average voting for each word, to define the sentiment of them. 5
- 6. APPROACH… 6
- 7. • Tokenizer: splits the text into very simple tokens • Sentence splitter: Fragments the text into its sentences • Gazetter: our Gazeteer is our gathered adjectives with their sentiment Approach… 7
- 8. APPROACH… • Jape Rules: To identify regular expressions we have formulated as grammar base ❖ Word level “”ﺑﯽ “Bi” and “”ﻧﺎ “Na” + noun => negative word (ﻣﻌﺮﻓﺖ ﺑﯽ ,)ﻧﺎﻣﺮد “ ”ن + Verb => negative verb (ﻓﮭﻤﺪ )ﻧﻤﯽ ❖ Sentence level “ He is not a lair” “That film had a lot of famous actors but it couldn’t attract people’s “ 8
- 9. TESTING • Performance test (1): Using a Persian web site : www.iran-booking.com 69% 9
- 10. 10
- 11. TESTING… • Performance test (2) Gathering about 5000 news from most famous Persian news websites depends on Alexa rating. Choose 800 news randomly and asked Persian people to vote on them about their sense. 11
- 12. PERFORMANCE TEST (2)… 1200 votes have been gathered on 800 news. 12
- 13. CONCLUSION AND FUTURE WORKS • For the first Persian API we receive to 69% accuracy till now How to improved the accuracy? ❖ more jape rules especially in the form of “BUT” “Although” and … ❖ Encouraging more Persian people to vote on the adjectives and phrases to have a better lexicon data base. ❖ Working on informal form too that’s common in tweets and comments. 13
- 14. REFERENCES • [1] A SVM-Based Method for Sentiment Analysis in Persian Language (Mohammad Sadegh Hajmohammadi, Roliana Ibrahim) • [2] https://gate.ac.uk/ • [3] “Majid Sezavar” http://sazvar.student.um.ac.ir/ • [4] http://www.coreservlets.com/Apache-Tomcat-Tutorial/tomcat-7-with -eclipse.html • [5] http://javapapers.com/web-service/java-web-service-using-eclipse 14
- 15. THANKS FOR YOUR ATTENTION… 15