Využití neuronových sítí při identifikaci objektů hodnocení v textu
| Název práce v češtině: | Využití neuronových sítí při identifikaci objektů hodnocení v textu |
|---|---|
| Název v anglickém jazyce: | Artificial Neural Networks for Opinion Target Identification |
| Klíčová slova: | neuronové sítě, hluboké učení, počítačové zpracování přirozeného jazyka, postojová analýza, hodnocení, identifikace objektů hodnocení v textu |
| Klíčová slova anglicky: | neural networks, deep learning, natural language processing, sentiment analysis, evaluation, target of evaluation, opinion target identification |
| Akademický rok vypsání: | 2023/2024 |
| Typ práce: | disertační práce |
| Jazyk práce: | |
| Ústav: | Ústav formální a aplikované lingvistiky (32-UFAL) |
| Vedoucí / školitel: | doc. Mgr. Barbora Vidová Hladká, Ph.D. |
| Řešitel: | |
| Konzultanti: | doc. Ing. Zdeněk Žabokrtský, Ph.D. |
| Zásady pro vypracování |
| Opinion target identification is in the long term one of the widely-discussed problems in the field of sentiment analysis. It has been explored using different approaches from the area of deep learning. However, the use of artificial neural networks in automatic opinion target identification, i.e. the task in which the evaluated entities need to be identified in natural language texts, has not been widely investigated. The student will make use of the existing data set for opinion target identification in different languages and compare the resulting neural networks with the results achieved by employing probabilistic models and, taking into account the relative language-independency of the given approach with state-of-the-art results achieved for various languages. The results of this research are further exploitable in other NLP tasks, such as e.g. automatic summarisation.
Expected Results: (1) Report on cross-lingual methods for sentiment analysis. (2) Workflows for employing artificial neural networks in sentiment analysis. (3) Language-independent opinion target identification system. |
| Seznam odborné literatury |
| - Pengfei Liu, Shafiq Joty, and Helen Meng. Fine-grained opinion mining with recurrent neural networks and word embeddings. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1433–1443, 2015.
- Wencan Luo and Diane Litman. Summarizing student responses to reflection prompts. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1955–1960, 2015. - Ying Ding, Jianfei Yu, and Jing Jiang. Recurrent neural networks with auxiliary labels for cross-domain opinion target extraction. In 31st AAAI Conference on Artificial Intelligence, 2017. |