Ananya Jha's repositories
App-Data-Scraping
This model aims at extracting metadata by hitting the google play-store API. This process has been accomplished using the playscrapper library.
ARIMA-Model-Time-Series-Forecasting
I have used ARIMA model for time-series forecasting. The dataset used in the implementation can be accessed from the data folder, alternatively you can also get the data here(https://raw.githubusercontent.com/jbrownlee/Datasets/master/shampoo.csv).
Bank-Churn-Survival-Analysis
We have predicted customer churn using the Bank Churn dataset collect from kaggle.
Basics-Regular-Expressions
Basics concepts of regular expressions(beginner level).
BeautifulSoup_Inspirational_Quotes
I have used BeautifulSoup library of python to scrape data from the web. The text file hence created contains inspirational quotes.
ChitChat-ChatBot
I have created this very basic rule based chat bot where I have used regular expression to design the rules.
CIFAR10_CNN
This is a model based on convolutional neural network. Here, the concept of pickling is used. Pickling is used to serialize
Comparison-of-various-tokenization-tools-in-NLP
This is a project based on
Digit-Recognition-Theano_and_Lasagne.
Testing git
EDA-of-Medical-Cost-Insurance-data---Kaggle
This project is exploratory data analysis of Medical Cost Personal Dataset acquired from Kaggle. The libraries involved to perform the analysis are - python's pandas, numpy, seaborn, matplotlib, missingno. Below are the steps summarizing the whole project : 1. Basic EDA : Shape of the dataframe(1338, 7), Statistical Analysis using describe function, columns in the dataset, memory used by the data set(73.3 + KB), Scatter plot and correlation matrix to look for correlation. 2. Missing Values : Library used missingno - generated matrix to look for missing values, isnull() method to cross check the presence of outliers in the data. 3. Duplicate values : Locations 195 and 581 have duplicate values, used .duplicated() method on the data frame. 4. Detecting Outliers : Scatter and boxplot to spot the outliers in the data(columns age, bmi and charges have outlier values). 5.Data analysis : Formed visualizations based upon questions - a. Mean value of children covered under insurance based on gender b. Average charges incurred by each age group(line plot). c. Mean body mass index based upon age(line plot) d. Are charges incurred different for smoker and non-smoker.(bar plot) e. People from which region incurred more medical costs.(bar plot) f. Which region has more number of people enrolling for the insurance scheme(bar plot) g. Were there more number of smokers who enrolled for insurance(bar plot) i. What is the gender of the people who generally smoke(bar plot)
False-News-Classification
This project we have used the knowledge of machine learning to classify false new. To achieve the purpose I have used passive aggressive classifier. To summarize the goodness of the model, I have used confusion matrix. This project is implemented using various python libraries on Google Collab.
HaarCascade-based-object-Detection
We have used open cv based haarcascade for object detection(face and eyes). This was developed by Viola Paul and Michael Jones. You can watch this video to get better insights. https://www.youtube.com/watch?v=88HdqNDQsEk
Image_Classification_Fashion_MNIST
This project involves classification of images using Convolutional Neural Architecture(CNN) using the keras framework of python.
Iris_KNN
This project is based upon the dataset collected from the kaggle repository(https://www.kaggle.com/uciml/iris/download).
Lead-Scoring
Lead Data has been pre-processed to formulate a final score based upon the data columns.
LIBRARY-MANAGEMENT-WEBSITE
This is a website on library management. This website has been established using html, css, bootstrap and javascript. SQLite has been used for the back end.
Logistic-Regression
I have performed Logistic Regression o n a dataset. The dataset is available in the data folder. I have referred-https://towardsdatascience.com/data-science-43c246d4eebc to this blog post
LSTM-1-Predicting-Stock-Prices
I have used LSTM network to predict stock prices. I have considered 10 years historical data for the purpose.
Parkinson_Disease_Detection
We have used XGBoost Classifier to accomplish the task of Parkinson's Disease Detection. The data used in the process is attached as 'parkinsons.data' file. This project is implemented on python framework using Google Collab.
Python
All Algorithms implemented in Python
Random-Forest-Classifier
This is an implementation of Random Forest Classifier.
Random-Forest-Regressor
We have used Random Forest Regressor Model to predict the target value petrol consumption in the dataset which has four independent variables
Recommendation-System
This is basic recommendation system based on metadata like popularity, weighted average and movie plot(content based).
Recurrent-Neural-Network
This repository contains material for learning Recurrent Neural Network.
Scraping-Mobile-Model-Name-to-Device-Name-Mapping-and-Price
Here, we have scraped google support site and pricebaba website to link mobile model name to mobile prices
Speech-To-Text-Part-1
This is a basic Speech to Text Conversion project. Python Libraries like speechrecognition and pyaudio are used. This is basically for learning purpose.
SQL-Interview
SQL interview questions
Web-Scraping-Website-Pricebaba-
We have scraped the website Pricebaba. We have created a program that takes price range value as input and returns price and VFM score of mobile phones in that range.