Used Exploratory Data Analysis(EDA) techniques,used pipeline based on Data Ingestion->Text Preprocessing(used word tokenization,stop words removal,lower casing,tags removal,etc)->Feature Engineering(used BOW,but other WorldtoVec deep learning technique we can use here as well,tf-idf,n-grams,one-hot encoding also can be use here as well)->Data Modelling->Data Evaluation Used different supervised learning algorihms under machine learning used for classification i.e. random forest classifier,decision tree etc Used my self created features,that helps to achieve better accuracy Evaluated it using confusion matrix,accuracy,and f1-score One can use stemming and lemmatization under text preprocessing also can use other ML algo and can use more data for analysis for better outcomes
Ansh2709/NLP-project
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|