Skip to content

A wide collection of various Machine learning applications including sentiment analysis, handwriting classification, recommendation engine and many more!!!!

Notifications You must be signed in to change notification settings

prankshaw/ML-Applications

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Machine Learning Application

Readme will be updated with complete description soon, website will be soon launched.
A wide collection of various Machine learning applications including sentiment analysis, handwriting classification, recommendation engine, color extraction, fashion recommender and identifier, Google API usage and many more!!!!

  • Some applications have been converted into web apps using Flask framework to simulate real world application of ML algorithms, and many more are yet to come.

  • Others can be interacted with, easily, by downloading the required file and having Jupyter Notebook installed on machine.

  • The google API usage files are complete code snippets which can be used just by copy-paste, with the only requuirement for an API key.
  • Description

    Amazon Apparel Product Recommendation System

    Applied Natural Language Feature Engineering and built a content based recommendation system using NLP Models (Bag of words and TF-IDF()TFIDF, short for term frequency–inverse document frequency, is a numerical statistic that is intended to reflect how important a word is to a document in a collection or corpus.) in problem of recommending similar products on e-commerce websites by using Amazon product advertising API.

    Tech Stack - Python, Bag of Words odel, TF-IDF, Word2Vec, NLP

    Handwritten Digit Recognition

    Applied K Nearest Neighbours Algorithm in recognizing handwritten digits (0 to 9) from the MNIST dataset. Developed a system from scratch which includes a machine to understand and classify the images of handwritten digits as 10 digits (0–9). Accuracy of the model was around 98%.

    Tech Stack - Python, ML(KNN), OpenCV, MNIST Database

    Dominant Color Extraction

    Applied K Means Clustering Unsupervised Algorithm in Segmenting an image into set ok K Dominant Colors. Extracted the K Dominant Colors from the image and then re-colorized it using those K Colors.

    Tech Stack - Python, ML(K-Means), OpenCV

    Spam or Ham Identification Web App

    Using Bag of words model we find the ‘term frequency’, i.e. number of occurrences of each word in the dataset, along with Term Frequency-Inverse Document Frequency to find probabality of each word, followed by additive smoothing.

    Tech Stack - Python, ML(Classification Algorithms)

    Tweets Sentiment Analysis

    It contains a basic use caseof Natural language processing, and using data classifiers on the dataset. Support vector machines are usedto effectively classify and accuracy is checked using F1 score and ROC-AUC curve.

    Tech Stack - Python, ML(Classification Algorithms)- SVM

    Movie Recommender System

    Fashion Accessories Identification

    Flower Species Detection Web App

    Credits

    The above code and application produced are just some of the Machine Learning applications that I worked on, through these years, and I have selected some of the relevant and intersesting applications for this repository, to help others learn and understand the real world applications of machine learning, so as to aid them to get out of theoretical knowledge of books. These work are develoed throughout my college period, while studying machine learning and deep learning, with help of lots of books, blogs, youtube videos and many more, who have helped me learn. Some of them might be inspired from these teachings, and will have some changes and learnings of my own, while most are original work, produced after deep understanding of the underlying technique. I will appreciate anybody who is willing to contribute or learn can just fork the repository and May the training parameters forever be in your favour :)
    Thank You!!

    Releases

    No releases published

    Packages

    No packages published

    Languages