tfidf-vectorizer

Objective of the repository is to learn and build machine learning models using Pytorch. 30DaysofML Using Pytorch

Document Search Engine project with TF-IDF abd Google universal sentence encoder model

The Bus-Mama is a bus tracking mobile application for the transportation of the students of BSMRSTU. It helps the students of our university by showing the available route, bus, and their exact location. This app includes real-time bus tracking which is going to solve a problem that university students have been facing for many years. Students are often seen missing their buses. Often they can't maintain the bus time. Since there are many buses in our university, students can easily catch a bus if they know where and when it will pass by. My goal is to track the buses and make hardware, mobile application, and machine learning solution to solve the issue. This way the students can get relief from missing the bus and use the buses efficiently. The main idea is to track the buses. GPS trackers will be attached to every bus that will give the current position of them and automatically sync on the server. The Bus-Mama mobile application will show every real-time position of those buses. This application will be installed on students' mobile phones and in this way the students can easily maintain their transportation. In this application, the current location of the bus can be seen through Google map. Every bus will have a specific marker on Google map and all the details about a specific bus will be shown by clicking on the marker. There will be seen about how far the bus is, from which direction it will come, how much time to reach the bus, how much time it will take if there is any traffic on road, etc. There is also a search option to know about any specific bus details. There is also a list of all buses with sufficient details that will help students to know about all the details. Every student will have an account through which they can access bus data. Another main objective is the Bus-Mama Chatbot in the Bengali language so that the students can communicate to know about the bus easily. For now, they can make conversation only about bus-related information. The Chatbot is not yet able to make conversation except bus-related questions. If anyone asks anything except bus-related questions, it cannot reply to the question rather it will give a tag to the question as a reply. As the Chatbot is created in the Bengali language, it has used the "trie" data structure in lemmatization. A library has been designed to lemmatize the Bengali words. Almost 63,205 Bengali words have been lemmatized by using the library to train the SVM machine learning model.

VIP Machine Learning Exercises and Practices

Scikit-learn tutorial for beginniers. How to perform classification, regression. How to measure machine learning model performacne acuuracy, presiccion, recall, ROC.

An NLP model to detect fake news and accurately classify a piece of news as REAL or FAKE trained on dataset provided by Kaggle.

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%.

📷🎥 Entity resolution system for SIGMOD 2020 programming contest

Authorship Attribution with Machine Learning

Fake News Detection System for detecting whether news is fake or not. The model is trained using "Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection. Link for dataset: https://arxiv.org/abs/1705.00648.

A text can be assigned more than one label

TFIDF being the most basic and simple topic in NLP, there's alot that can be done using TFIDF only! So, in this repo, I'll be adding the blog, TFIDF basics, wonders done using tfidf etc.

Performed text preprocessing, clustering and analyzed the data from different books using K-means, EM, Hierarchical clustering algorithms and calculated Kappa, Consistency, Cohesion or Silhouette for the same.

Detect Real or Fake News. To build a model to accurately classify a piece of news as REAL or FAKE. Using sklearn, build a TfidfVectorizer on the provided dataset. Then, initialize a PassiveAggressive Classifier and fit the model. In the end, the accuracy score and the confusion matrix tell us how well our model fares.

Practice Rust by making Vectorizer

This notebook contains entire text preprocessing pipeline for NLP problems. The ready-to-use functions require NLTK and SKlearn package installations. It also contains some prominent text classification models.

Goal is to classify the given text under two categories:- Label 1 and Label 2.

This competition is hosted by Kaggle https://www.kaggle.com/c/nlp-getting-started/overview. I participated in the competition in order to try my hands on the field of Artificial Intelligence known as Natural Language Processing.

Different NLP Models on Kaggle Dataset

Classify fake news and real news using three machine techniques. We are using tfidf vectorizer to vctorize the newsbody.

Repositorio com códigos relacionados a pesquisa de TCC sobre desempenho dos algoritmos Naive Bayes, RL e SVM para classificação de revisões.

This repository hosts the programming exercises for the course Neural Networks of NTUA ECE.

Multi label classification. Dataset is from Kaggle(https://www.kaggle.com/badalgupta/stack-overflow-tag-prediction)

Predict tags on StackOverflow with linear models - Week 1 assignment of Coursera's Natural Language Processing course from the Advanced Machine Learning Specialization.

A NLP project about a brief application of tfidf vectorizer for text feature extraction and pipeline and feature union to chain multiple steps into one

Public documentation for some of my Data Science related projects

Predicting the value of jeopardy questions based on each questions’ text and answer.