This is a repository containing models submitted as part of our participation in SemEval Task 6 OffensEval: Identifying and categorizing Offensive lnaguage in social media by SemEval 2019 by (Zampieri et al.,2019b).
- Preprocess.py Some of basic Preprocessing as Step 1 to clean the data.
- TaskA.py: It is a classification model submitted in Task A dealing with 2 classes.
- TaskB.py: This model submitted as Task B is utilizing both the word and pos features. At first POS Tag for all the sentences can be get by POS_Tag.py in form of text file and then inserted at line 50.
- TaskC.py : This is for the third task i.e Target Identification
- POS_Tag.py: This can be used to get the POS tag for all the sentences using nltk.
Python:2.7.5 numpy: 1.16.0 gensim: 3.7.0 tensorflow: 1.14.0 nltk:3.4 pandas: 0.23.4
https://www.aclweb.org/anthology/S19-2105.pdf
@inproceedings{kapil2019nlp, title={NLP at SemEval-2019 Task 6: Detecting Offensive language using Neural Networks}, author={Kapil, Prashant and Ekbal, Asif and Das, Dipankar}, booktitle={Proceedings of the 13th International Workshop on Semantic Evaluation}, pages={587--592}, year={2019} }