Mercari Price Suggestion

Last Updated on May 3, 2021


This project is for predicting prices for products based on some features. I have written a blog on medium you can check it -

Github link --

More Details: Mercari price suggestion

Submitted By

Share with someone who needs it

Convolutional Neural Network Application To Classify Identical Twins(December 2019-March 2020) (Python,Ml)

Last Updated on May 3, 2021


-Transfer learning is employed to classify the images of identical twins.

-Transfer learning basically means using the knowledge acquired during a previous problem solving for finding a solution to a new problem in hand.

-Weights and biases values are used with some fine tuning.

-Pre-trained Convolutional Neural Network models which are developed for image recognition tasks on ImageNet data set, provided by Keras API is used as feature extractor preprocessor.

-All these models are proven to be very efficient in image recognition task.They have showcased very high accuracy on ImageNet dataset.The labels they have given for images were highly accurate with very less error percentage.

-Only convolutional base layers are used here,that is fully connected layers are not included here.

-Fully connected layer is not included and a new fully connected layer is addee at the end for the required categorisation of the data.

-During dataset building-collected images of the identical twins.Two categoried were defined in this way.

-VGG19 is used as a standalone program to extract features from the dataset.

-Feature vectors for training dataset is obtained and mean feature vector for both categories were calculated.

-Testing is done by comparing the feature vectors of testing data with mean feature vectors of each category using cosine similarity.

-Obtained a fair accuracy while testing. 

More Details: Convolutional Neural Network Application to classify Identical Twins(December 2019-March 2020) (Python,ML)

Submitted By

Book Recommendation System

Last Updated on May 3, 2021


Book Recommendation System

Book recommendation is created and deployed in this approach of work, which helps in recommending books. Recommendation achieved by the users feedbacks and rating, this is the online which analyse the ratings, comments and reviews of user, negative positive nature of comments using opinion mining. User searching for the interested book will be displayed in top list and also can read feedback given by people about the book or any searched items. Whenever user search for any book from the large data available, he gets confused from the number of displayed item, which one to choose. In that case recommendation helps and displays on the interested items. This is the trustworthy approach, which is used in this project where selection is based on the dataset.


This project used clustering as the central idea. A clustering approach is used. Clustering is based on similarity where similar elements are kept in a single group. Likewise similar element, the irrelevant elements are also reside in a group, which is another group, based on similarity value or maximum size of cluster. The clustering approach which is used in our work is K-mean clustering for grouping of similar users. It is the unsupervised and simplest learning algorithm, which simplifies mining work by grouping similar elements forming cluster. This is done using a parameter called K-centroids. Distance between each element is calculated for checking the similarity and forming a single cluster to reside the similar elements, after comparing with K-centroid parameter.

In this project, 6 clusters were made.

The project is made with 2 separate datsets in .csv format taken from Kaggle.

  1. Books dataset
  2. Ratings

This project is GUI based. The output page has 2 options:

  1. Rate books
  2. Recommend books

The user can chose either according to themselves.

Rate books

In this option, the user can rate books.

Recommend books

In this option the books are recommended to the user, according to their previous readings.

More Details: Book Recommendation System

Submitted By

Fantasy Cricket Game

Last Updated on May 3, 2021


It is an online game where you create a virtual team of real cricket players and score points depending on how your chosen players perform in real life matches. To win a tournament, you must try and get the maximum points and the No. 1 rank amongst other participants. Here's how a Fantasy Cricket game may look like. 


1 Opening screen of the application. You can see the players of each category by selecting the category. To begin with, the selection is disabled until a new team is created from the Manage Teams menu. A pop up asking the name of the team appears. 

2 The toolbar menu options which allow you to create a new team, open an existing team, save your team and finally evaluate the score of a saved team.  



3 After clicking create Team, the left box is populated with player names. As you select a different category, the corresponding list of players is displayed.


4 On double-clicking each player name, the right box gets populated. Points available and used are displayed accordingly.



5 Message if the game logic is not followed 

6 Pop-up on clicking Evaluate Score. You can select your team here and the match for which the players' performance is compared. 

7 The final score for your fantasy team based on the match selected.  

More Details: Fantasy Cricket Game

Submitted By

Machine Learning Algorithms

Last Updated on May 3, 2021


I have created this projects by learning some machine learning algorithm's

Different algorithms I have learned are :

  1. K-means : In k-means algorithm I learned the working of algorithm using a dataset and loaded it. I have used sklearn and used metrics function in it to predict best fit score for dataset.
  2. KNN : In this algorithm I have used a csv file in order to perform operations on it. I have trained the data and then labels of columns in order to fit the data in model. Then have predicted the acurracy of model.
  3. Regression : In this algorithm I have used different libraries such as numpy, pandas, sklearn and matplotlib for working on excel file. numpy and pandas is used for reading the csv file form directory and also to use series and dataframes of pandas. Matplotlib is used for graphical representation of model and sklearn is used to import its linear model for the data and train the data.
  4. SVM(Support Vector Machine Algorithm): In this algorithm I have used a different data set. Loaded that dataset into the algorithm with help of sklearn. works for classification and for regression, svm uses hyperplay to divide data in straights(line, 4D). Its a linear way to divide data.
More Details: Machine learning algorithms

Submitted By

Emotional Analysis Based Content Recommendation System

Last Updated on May 3, 2021


As the saying goes, “We are what we see”; the content we see may have an adverse effect on our behavior sometimes. Especially in a country like India, where numerous films and TV series are highly prominent, there are great chances of watching explicit or disturbing content randomly. This may have adverse effects on behavior of people, especially children. And we also know “Prevention is better than cure”. Preventing inappropriate content from going online can be more effective than banning them after release.

To achieve this, we aim to create a content filtering and recommendation system that either recommends a film or TV series or alerts a user with a warning message saying it’s not recommended to watch. Netflix or any other Over-the-top (OTT) platforms perform a filtering process before they buy digital rights for any content. This is where our tool comes handy. It detects absurd or hard emotion inducing content with the help of human emotions. Through this project we aim to create a content detector based on human emotion recognition. We will project scenes to test audience and capture their live emotions.

Then we use “Facebook Deep Face”, a pre-defined CNN based face recognition and facial emotion analysis model to identify faces and analyze their emotions. We use “Deep Learning” methods to recognize facial expressions and then make use of Circumplex Model proposed by James Russell to classify emotions based on arousal and valence values. Based on majority emotion that is projected by audience we would either recommend or not recommend the content for going on-air. This system prevents inappropriate content from going on-air

More Details: Emotional Analysis Based content recommendation system

Submitted By