재난 주제 소셜미디어 데이터셋 구축을 위한 주제 적합성 분류 모델 연구
An ensemble machine learning model for identifying relevant messages for social media
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
소셜미디어에서 매일 생산되고 유통되는 수많은 정보는 데이터의 과잉으로이어져 필요한 때에 필요한 정보를 찾는 것을 어렵게 만든다. 본 연구는 이러한데이터에서 필요한 정보를 자동 추출할 수 있는 앙상블 기반의 주제 적합성 분류 모델을 제안하였다. 이 모델은 소셜미디어의 텍스트에 원하는 정보가 있는지혹은 텍스트가 주어진 주제에 적합한지 등을 자동으로 분류하는 모델이다. 본연구에서는 3개의 기본 모델로 구성된 앙상블 모델을 설계한 후 모델의 학습단계에서 실제 데이터에서의 클래스 분포를 반영할 수 있도록 학습 데이터를공급하는 배치 생성기를 활용하고, 추론 단계에서는 출력값 임계치를 시기별로조정하는 임계치 이동 기법을 사용하였다. 이 모델은 소셜미디어를 비즈니스, 정책, 마케팅, 위기 관리 등 다양한 분야에서 활용할 수 있게 해 줄 수 있다. 또한 감정 분석이나 태도 분석, 거짓 정보 탐색 등의 작업에 앞선 일종의 전처리처리 모델로 사용할 수도 있다.
Abstract
The excessive amount of information produced and circulated daily on social media has made it challenging to find relevant information when needed. This study proposes an ensemble-based topic relevance classification model that can automatically extract necessary information from such data. The model automatically classifies whether the text on social media contains desired information or is relevant to a given topic. This model automatically classifies whether the text on social media contains desired information or is relevant to a given topic. In this study, an ensemble model consisting of three base models is designed. During the model's training phase, a batch generator is used to supply training data that reflects the class distribution in actual data. In the inference phase, a threshold-shifting technique is applied to adjust output thresholds. This model can be applied in various fields such as business, policy-making, marketing, crisis management, etc., making social media a valuable resource. Additionally, it can serve as a preprocessing model for tasks like sentiment analysis, attitude analysis, and fake information detection.
참고문헌 (36)
- 이준범, "KcBERT: 한국어 댓글로 학습한 BERT", 제 32 회 한글 및 한국어정보처리 학술대회 논문집, pp.437~440 (2020)
- 최지명, "기계학습 알고리즘을 이용한 한국어 텍스트 저자 판별: 블로그의 영화리뷰를 대상으로", 석사, pp.2015
- 최지명, "기계학습을 이용한 역사 텍스트의 저자판별: 1920년대 개벽 잡지의 논설 텍스트", 언어와 정보, 22(1), pp.91~122 (2018), 한국언어정보학회[DOI]
- Abdul-Mageed, M., "Mega-COV: A billion-scale dataset of 100+ languages for COVID-19", ~ (2021)
- Acheampong, F. A., "Transformer models for text-based emotion detection: a review of BERT-based approaches", Artificial Intelligence Review, pp.1~41 (2021)
- Biggers, F. B., "A deep semantic matching approach for identifying relevant messages for social media analysis", Scientific Reports, 13(1), pp.12005~ (2023)
- Chen, E., "Tracking social media discourse about the covid-19 pandemic: Development of a public coronavirus twitter data set", JMIR public health and surveillance, 6(2), pp.e19273~ (2020)
- Cobo, A., "Identifying relevant messages in a twitter-based citizen channel for natural disaster situations", Proceedings of the 24th international conference on world wide web, pp.1189~1194 (2015)
- Devlin, J., "Bert: Pre-training of deep bidirectional transformers for language understanding", Proceedings of naacL-HLT, 1(2) (2019)
- Feng, Y., "Short text sentiment analysis based on multi-channel CNN with multi-head attention mechanism", IEEE Access, 9, pp.19854~19863 (2021)
- Ghatasheh, N., "Modified genetic algorithm for feature selection and hyper parameter optimization: case of XGBoost in spam prediction", IEEE Access, 10, pp.84365~84383 (2022)
- Hamidian, S., "Rumor detection and classification for twitter data", ~ (2019)
- Hassan, S. U., "Analytics of machine learning-based algorithms for text classification", Sustainable Operations and Computers, 3, pp.238~248 (2022)
- Hayawi, K., "ANTi-Vax: a novel Twitter dataset for COVID-19 vaccine misinformation detection", Public health, 203, pp.23~30 (2022)
- Kaufhold, M. A., "Rapid relevance classification of social media posts in disasters and emergencies: A system and evaluation featuring active, incremental and online learning", Information Processing & Management, 57(1), pp.102132~ (2020)
- Khaleq, A. A., "Twitter analytics for disaster relevance and disaster phase discovery", Proceedings of the Future Technologies Conference (FTC) 2018, 1, pp.401~417 (2019), Springer International Publishing
- Kim, Y., "Convolutional neural networks for sentence classification", ~ (2014)
- Kimura, Y., "Overview of the NTCIR-14 QA Lab-PoliInfo task", Proceedings of the 14th NTCIR Conference, 14, pp.121~140 (2019)
- Kumar, P. Monika, "Improved Tweet Sentiment Analysis by Features Weight Optimize by GWO and Classify by XG-Boost", Sentiment Analysis and Deep Learning: Proceedings of ICSADL2022, pp.607~614 (2023), Springer Nature Singapore
- Liu, Z., "Multichannel cnn with attention for text classification", ~ (2020)
- Lu, G., "CNN-BiLSTM-Attention: A multi-label neural classifier for short texts with a small set of labels", Information Processing & Management, 60(3), pp.103320~ (2023)
- Mukhtiar, W., "Relevance Classification of Flood-related Twitter Posts via Multiple Transformers", ~ (2023)
- Palanivinayagam, A., "Twenty Years of Machine-Learning-Based Text Classification: A Systematic Review", Algorithms, 16(5), pp.236~ (2023)
- Paul, S., "CyberBERT: BERT for cyberbullying identification: BERT for cyberbullying identification", Multimedia Systems, 28(6), pp.1897~1904 (2022)
- Ramteke, J., "Election result prediction using Twitter sentiment analysis", 2016 international conference on inventive computation technologies (ICICT), 1, pp.1~5 (2016)
- Sawhney, R., "BMGC 2020 Grand Challenge: Multi-Aspect Analysis of the MeToo Movement on Twitter", 2020 IEEE Sixth International Conference on Multimedia Big Data (BigMM), pp.481~484 (2020), IEEE
- Shi, P., "Simple bert models for relation extraction and semantic role labeling", ~ (2019)
- Shirafuji, D., "STARS at the NTCIR-14 QA Lab-PoliInfo Classification Task", Proceedings of the 14th NTCIR Conference on Evaluation of Information Access Technologies, pp.10~13 (2019)
- Singla, J, "Comparing ROC curve based thresholding methods in online transactions fraud detection system using deep learning", 2021 International Conference on Computing, Communication, and Intelligent systems, pp.9~12 (2021)
- Vera-Burgos, C. M., "Using Twitter for crisis communications in a natural disaster: Hurricane Harvey", Heliyon, 6(9), pp.1~10 (2020)
- Wang, L., "Multichannel Two-Dimensional Convolutional Neural Network Based on Interactive Features and Group Strategy for Chinese Sentiment Analysis", Sensors, 22(3), pp.714~ (2022)
- Wei, H., "News-Comment Relevance Classification Algorithm Based on Feature Extraction", 2021 International Conference on Big Data Analysis and Computer Science (BDACS), pp.149~152 (2021)
- Xie, J., "Self-attention-based BiLSTM model for short text fine-grained sentiment classification", IEEE Access, 7, pp.180558~180570 (2019)
- Yong, J, "A Cross-Topic Method for Supervised Relevance Classification", Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019), pp.147~152 (2019)
- Zhang, X., "Cyberbullying detection with a pronunciation based convolutional neural network", 2016 15th IEEE international conference on machine learning and applications (ICMLA), pp.740~745 (2016)
- Zhou, Z. H., "Training cost-sensitive neural networks with methods addressing the class imbalance problem", IEEE Transactions on Knowledge and Data Engineering, 18(1), pp.63~77 (2005)