ghoshal7 Goto Github PK

bank_check_segmentation

capacitated_vehicle_routing_problem

Capacitated Vehicle Routing Problem

cfpb_complaint_classification

chicago-taxi-revenue

The objective of this project is to build visualization models which help in the interpretation of how Chicagoan uses their public taxi on an hourly and daily basis

deep_learning-neural_network_1

econometric-time-series-analysis-of-stock-price

engineering_data_analysis

fake_or_real_news

gpr_database

handwriting_recognition

lp-using-cplex

maximum_independent_set-mis-

movieclustering

What makes a good movie? Most of the top-rated movies in the International movie database (IMDB) are critically acclaimed and are generally a safe bet in terms of commercial success. Naturally, it would be interesting to investigate if these top movies have some distinct features responsible for their high ratings. This project aims to find out the type of natural cluster that exists among the top 250 movies from IMDB. Unsupervised machine learning techniques will be employed, more specifically, clustering algorithms. Hopefully, these clusters will give us information to observe the recurrent pattern. To build our dataset we used OMDB's web API which is RESTful web service to obtain movie information. For plot summaries we scrapped a movie's plot from IMDB's website using BeautifulSoup Library in python. Our final extracted dataframe had 250 rows and 113 columns. The inputs to our dataframe were all categorical features which were one hot encoded. We first used dimensionality reduction techniques such as PCA which was followed by K-Means and DB-SCAN clustering to find inherent clusters in the data.

pandas_exercises

Practice your pandas skills!

pdf_extraction

practice-stuffs

python_basics

"Basics of Python" - Data Analytics Club @ TAMU

summary-of-repositories

supply_chain_analytics

toxic_comment-classification-with-rnn

Using Keras implemented RNN to classify Text

ucn

wind-power-prediction