| _version_ | 1866901950613684224 |
|---|---|
| author | Mustafa Demir* Yakup Kutlu |
| author_facet | Mustafa Demir* Yakup Kutlu |
| contents | The use of social media has been continuously increasing, and our comments and posts on various topics have now reached massive proportions. The analysis of these shared data has become significantly important, as recent studies indicate. Evaluating and interpreting social media content enables necessary measures to be taken. In particular, analysing social media content is crucial for brands to communicate more effectively with their target audiences. This analysis helps to understand user trends, interests, and feedback, thereby supporting the optimization of content strategies. It also provides valuable insights for competitive analysis and market positioning, allowing brands to run more effective campaigns and improve customer satisfaction. In this study, the impact of pre-processing tools like Turkish Stemmer and Zemberek, commonly used in Turkish texts, on performance was examined. Results show that pre-processing significantly enhances performance in Turkish texts. |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_14587317 |
| institution | Zenodo |
| language | eng |
| publishDate | 2021 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | The Effect of Preprocessing Stage on Sentiment Analysis in Turkish Texts Mustafa Demir* Yakup Kutlu Social media Classification Sentiment Analysis The use of social media has been continuously increasing, and our comments and posts on various topics have now reached massive proportions. The analysis of these shared data has become significantly important, as recent studies indicate. Evaluating and interpreting social media content enables necessary measures to be taken. In particular, analysing social media content is crucial for brands to communicate more effectively with their target audiences. This analysis helps to understand user trends, interests, and feedback, thereby supporting the optimization of content strategies. It also provides valuable insights for competitive analysis and market positioning, allowing brands to run more effective campaigns and improve customer satisfaction. In this study, the impact of pre-processing tools like Turkish Stemmer and Zemberek, commonly used in Turkish texts, on performance was examined. Results show that pre-processing significantly enhances performance in Turkish texts. |
| title | The Effect of Preprocessing Stage on Sentiment Analysis in Turkish Texts |
| topic | Social media Classification Sentiment Analysis |
| url | https://doi.org/10.5281/zenodo.14587317 |