complaint-triage
Multi-class text classifier that routes consumer complaints to the right product category using TF-IDF, LinearSVC and a CLI router.
- 45
- Jupyter Notebook
- 3 wk ago
Multi-class text classifier that routes consumer complaints to the right product category using TF-IDF, LinearSVC and a CLI router.
A Natural Language Processing with SMS Data to predict whether the SMS is Spam/Ham with various ML Algorithms like multinomial-naive-bayes,logistic regression,svm,decision trees to compare accuracy and using various data cleaning and processing techniques like PorterStemmer,CountVectorizer,TFIDF Vetorizer,WordnetLemmatizer. It is implemented using LSTM and Word Embeddings to gain accuracy of 97.84
Spam SMS Detection Project implemented using NLP & Transformers. DistilBERT - a hugging face Transformer model for text classification is used to fine-tune to best suit data to achieve the best results. Multinomial Naive Bayes achieved an F1 score of 0.94, the model was deployed on the Flask server. Application deployed in Google Cloud Platform
A machine learning model that predicts tags for a given question and body.
Perform sentimental analysis on the Elon-musk tweets (Elon-musk.csv)
The project utilizes the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm. The main objective of this project is to measure the similarity between text documents using the TF-IDF algorithm.
This is the advanced form of the sentiment analysis from text. In this method we have to find out the sentiment regarding the aspect word. The data set contains the reviews and the aspect word/phrase. We have to find out the sentiment regarding that aspect/phrase. We have used simple method using TfIdfVectorizer and ml algorithms. Got the nearly 70% accuracy.
Click below to checkout the website of this ML-NLP Project
Classify the given genetic variations/mutations based on evidence from text-based clinical literature
Accurate 6-emotion detection from text (93% accuracy). Powered by GloVe embeddings and Bidirectional LSTM. Real-time predictions via Streamlit for rapid insights into sentiments.