shakhawathassan Goto Github PK

american_cities_crime_rate_case_study

The purpose of this research is to see if the factors such as number of police, education level of the civilian population, unemployment rate, income per capita, and ethnic composition of population influence crime rate per 100,000 population. For variance analysis, both Heteroscedasticity Test and White’s General heteroscedasticity Test scattered on EViews. Based on data analysis, the result shows that about 33% of adults who have higher degrees are less likely to commit a crime than those who don’t have the higher degrees.

cipherkey_encrypted_messege

citi_bike_communicate_data_findings

Developed statistical models with data visualizations to provide solution recommendations for the client. Performed data cleaning using Pandas and NumPy to ensure the correctness and completeness of the dataset and create effective visualizations to help the audience derive greater insights

citi_bike_data_analysis_march20

client-s_website_a-b_test_results

Helped a client to understand if they should keep the implemented new page or keep the old page.

covid-19_effects_on_job_economy

data_scientist_vs_data_analyst_job_market_analysis

We analyzed the job market for data analysts and data scientists based on their avg estimated wages and degree requirements, company type, and the industry type. Excel used for data cleaning and charts and Tableau used for visualizing.

e-commerce_website_testing

This project was done by developing 10 user stories from the application, writing test cases for the user stories using Gherkins, and developing a test framework using the following technology such as Maven, Git, TestNG, Cucumber, Page Object Model (POM).And JIRA was used as an issue tracking tool. And this project was created by following the SDLC methodology and JAVA

linux_python_scripts

nyc_vs_global_weather_trends

This project is comparing the global temperature and New York City's temperature. As the global temperature rises, NYC's temperature rises as well. The data was collected from the Kaggle data source by using queries. Later, charts and tables were done in Excel.

tmdb-movies_data_analysis

In the first section, I pulled up the tmdb datasets. Next, then I cleaned the data, fixed the data type, split the rows which carry "|", remove null values, and remove duplicate rows. Before cleaning the datasets, I had 10,866 rows and 21 columns. After cleaning the datasets, In the exploratory section, I looked for the findings. For instance, action, adventure, drama, comedy and thriller genres are most popular from year to year (1960-2015). Action and adventure take the highest place for genres. Avatar is the highest grossing movie of all time (rev 2.78 Billion). Furthermore, The mean revenue was about 107 Million and the highest budget of a film was about 425 million. As the year goes on, the number of movies is increasing. On the same hand, genres, casts, directors are the main properties which heavily depend on a film's revenue. For example, Avatar is an action packed film which is directed by James Cameron, and it's the highest grossing film of all time (as of 2015). Top 3 directors are Steven Spielberg, James Cameron and Peter Jackson. Steven Spielberg is the number one director among all of them. Moreover, Harrison Ford, Tom Cruise, and Tom Hanks are the ones who have gotten the highest revenues. Lastly, there hasn't been any correlation found between budget and revenue, and revenue and runtime.

weratedogs_data_wrangling

Data Wrangling on WeRateDogs (Tweeter account). After the data analysis process, it shows that there are more favorite tweets than counts of a retweet on WeRateDogs. This means people are favoriting a picture of a dog than they are retweeting that tweet.