sumitm0432 Goto Github PK

adult-census-income-predictions-and-eda

This data was extracted from the 1994 Census bureau database by Ronny Kohavi and Barry Becker (Data Mining and Visualization, Silicon Graphics). A set of reasonably clean records was extracted using the following conditions: ((AAGE>16) && (AGI>100) && (AFNLWGT>1) && (HRSWK>0)). The prediction task is to determine whether a person makes over $50K a year and the exploratory data analysis.

aits-iris-eda-and-clustering

Exploratory data analysis and Cluster Algorithms (AITS Assignment)

amazon-topical-chat-dataset-text-classification

EDA and text classification of amazon topical chat dataset

article-summarizer-using-nlp

Used the Natural Language Processing to summarize the articles based on sentence ranking. The summary of the articles is an extractive summary.

black-friday-sales-predictions

A retail company wants to understand the customer purchase behavior (specifically, purchase amount) against various products of different categories. They have shared purchase summary of various customers for selected high-volume products from last month. The data set also contains customer demographics like age, Gender, Marital Status etc. and Total purchase amount from last month.

drug-review-sentiment-analysis-and-eda

The objective of this project is to predict the sentiment of the drug Users, according to their reviews and various other features like the condition they are suffering from, the rating of the drug used, Date of the usage, and others. Exploratory Data Analysis is done to get the insights and Feature engineering is done. Machine learning models are developed for the prediction of the sentiment and Feature importance is plotted.

explicit-content-classifier-using-resnet

The objective of this project is to classify explicit content that contains inappropriate images like pornography and Hentai. The classifier used for this is ResNet50 and ResNet101 also known as Residual Neural Network. There are five categories that the model is trained on which are Porn, Hentai, Sexy, Drawing, and Neutral. Porn, Hentai, and Sexy can be classified as NSFW (Not Safe For Work) further and the other two are SFW (Safe For Work).

forest-fire-area-prediction-and-eda

It's the Project, me and my team of Tanmay and Simrann made at the UHack 4.0 (USIT), a 24 hour hackathon.

hospital-reviews-and-rating-extraction-using-google-places-and-geocoding-api.

The Hospital Reviews and Ratings Extraction is done using the Google Place API and Google Geocoding API.

hotel-booking-cancellation-prediction-and-sa

The objective of the project is to predict the hotel booking status of the guest if it'll be canceled or not based on the various features like ADR (Average Daily Rate), booking changes, lead time, type of the hotel booked, and more. The Exploratory Data Analysis and Statistical Analysis is done for insights and feature engineering. Four Machine learning and Deep Learning Models are trained for that purpose.

internshowcase

Open Source Social Projects from AITS associates with 💝

malicious-webpage-classifier

The objective of this project is to classify the web pages into two categories Malicious[Bad] and Benign[Good] webpages. Exploratory Data Analysis and Geospatial Data Analysis are done to get more insights and knowledge about the data. Features are engineered and the data is preprocessed accordingly. A total of four ML and DL models are trained. The models are XGBoost, Logistic Regression, Decision Tree and Deep Neural Network. The DNN is implemented in PyTorch and the others are implemented using scikit learn.

pushshift-flair-dataset-extraction

quora-insincere-questions-classification

Quora Insincere Questions Classification Project Under Dr. Sri Phani Krishna Karri at National Institute of Technology, Andhra Pradesh.

textblob

Simple, Pythonic, text processing--Sentiment analysis, part-of-speech tagging, noun phrase extraction, translation, and more.

tweet-sentiment-extraction-using-bert

Using BERT [pytorch] to extract the sentiment of a tweet which is the smallest text that represents the sentiment of the tweet.

twitter-sentiment-analysis

The objective of this task is to detect hate speech in tweets. We say a tweet contains hate speech if it has a racist or sexist sentiment associated with it. So, the task is to classify racist or sexist tweets from other tweets.

xlm-roberta-for-textual-entailment

A multilingual model XLM- RoBERTa for the textual entailment of sequence pair - premise and hypothesis of 15 different languages using the MNLI and XNLI corpus.