Experimental Study of Methods

1.

Tikhonov Moscow Institute of Electronics
and Mathematics
The Department of Computer
Engeneering
Experimental Study of Methods for
Semantic Text Analysis
Student: Karina Malyshkina, BIV 206
Program: Information Science and Computation Technology
Academic Advisor: S. A. Slastnikov, Ph.D. in Technology
Moscow 2024

2.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Content plan
• Introduction and key terms
• Research background
• Problem statement
• Research aims
• Research design
• Results
• References
Content plan
2

3.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Introduction and key terms
3
Introduction and key terms
Natural Language Processing
Named Entity Recognition
Sentiment Analysis

4.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Research background
Research Background
Some existing studies on similar topics
4

5.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
5
Problem Statement
Problem Statement
Result of excerpt sentiment analysis
Excerpt from a news article about the Oscar

6.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Research aims
6
NER
Sentiment
Аль Пачино
negative
Джессика Лэнг
neutral
Арнольд Шварценеггер
positive
Николас Кейдж
positive
Research aims
Result of each NER sentiment analysis
Excerpt with highlighted NER

7.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Research design
7
Research design
Flowchart of sequential work stages
Achieved results
Anticipated results

8.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Achieved results
8
Achieved results. Text’s segments detection
Problems in Russian language texts:
1) Number
2) Case
3) Gender
Solving:
• Using ensemble of Pymorphy and Spacy libraries
The result of the second part of the program
Detected text segments in Oscar excerpt

9.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
Anticipated results
Experimental ranking of libraries for
sentiment analysis:
Natural Language Toolkit (NLTK)
TextBlob
VADER
DeepPavlov
Sentiment_tool library
9
Anticipated results
Chart result of sentiment analysis for different NERs
Аль Пачино
Николас
Кейдж
Арнольд
Шварцнегер
Джессика
Лэнг
0
1
2
Negative
3
Positive
4
Neutral
5
6

10.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
References
10
References
[1] A. Ushio and J. Camacho-Collados, "T-NER: An All-Round Python Library for Transformer-based Named Entity Recognition," Proceedings of the 16th
Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021): System Demonstrations, 2021. [Online]. Available:
https://doi.org/10.48550/arXiv.2209.12616.
[2] H. Shelar, G. Kaur, N. Heda and P. Agrawal, "Named Entity Recognition Approaches and Their Comparison for Custom NER Model," Science & Technology
Libraries, vol. 39, no. 3, pp. 324-337, 2020. [Online]. Available: https://doi.org/10.1080/0194262X.2020.1759479.
[3] C. Kaur and A. Sharma, "Social Issues Sentiment Analysis using Python," 2020 5th International Conference on Computing, Communication and Security
(ICCCS), 2020. DOI: 10.1109/ICCCS49678.2020.9277251.
[4] S. Siddharth, R. Darsini, and M. Sujithra, "Sentiment Analysis on Twitter Data Using Machine Learning Algorithms in Python," Conference Paper,
Coimbatore Institute of Technology, Coimbatore, 2018.
[5] W. Medhat, A. Hassan, and H. Korashy, "Sentiment analysis algorithms and applications: A survey," Ain Shams Engineering Journal, vol. 5, no. 4, pp. 10931113, 2014. [Online]. Available: https://doi.org/10.1016/j.asej.2014.04.011
[6] C. Kaur and A. Sharma, "Twitter Sentiment Analysis on Coronavirus using Textblob," EasyChair Preprint, no. 2974, 2020.
[7] S. Oswal, R. Soni, O. Narvekar, and A. Pradha, "Named Entity Recognition and Aspect based Sentiment Analysis," International Journal of Computer
Applications, vol. 178, no. 46, 2019.
[8] L. Nemes and A. Kiss, "Information Extraction and Named Entity Recognition Supported Social Media Sentiment Analysis during the COVID-19 Pandemic,"
Appl. Sci., vol. 11, no. 22, 11017, 2021. [Online]. Available: https://doi.org/10.3390/app112211017
[9] L. Zhao, L. Li, X. Zheng, and J. Zhang, "A BERT based Sentiment Analysis and Key Entity Detection Approach for Online Financial Texts," 2021 IEEE 24th
International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2021, pp. 1-6. DOI:10.1109/CSCWD49262.2021.9437616
[10] A.N. Tarasova and K.O. Ivanov, "Сентиментальный анализ постов в социальных сетях посредством Python [Sentimental Analysis of Posts in Social
Networks Using Python]," International Scientific Journal "Symbol of Science," no. 3-1, pp. 10-12, 2022.
[11] OTUS Blog, "Обработка и анализ естественного языка с помощью Python-библиотеки spaCy [Processing and Analysis of Natural Language Using
Python spaCy Library]," 2023. [Online]. Available: https://habr.com/ru/companies/otus/articles/755584/
[12] D. Zagorodnev, "NLP и аудит [NLP and Audit]," 2021. [Online]. Available: https://newtechaudit.ru/nlp-i-audit/

11.

12.

Tikhonov Moscow Institute
of Electronics and Mathematics
Experimental Study of Methods for
Semantic Text Analysis
12
Achieved results
Achieved results. NER detection
The result of the first part of the program
Detected NER in Oscar excerpt
Top-5 libraries for NER detection
1. Natasha 2. spaCy 3. SlovNet 4. DeepPavlov 5. RuBert Transformer