I analyzed 5,000 tweets from Twitter user @dog_rates. This project aims to show expertise in using Python to gather data from a variety of sources; querying Twitter's API with the Tweepy library and also using the Request library to programmatically get a file from the internet.
The dataset for this analysis contains daily & cumulative number of COVID-19 tests conducted, number of positive, hospitalized, recovered & death cases reported by country. R programming language and the TidyVerse package were used to clean and analyze the data.
Analysis and visualization of 54,000 diamonds to determine pointers to their pricing based on characteristics such as Carat, Cut, Clarity, and Color. This project uses both the Matplotlib and Seaborn libraries and includes both Univariate and Multivariate charts depending on whether the variables are Quantitative or Categorical.
Over 100,000 medical appointments in Brazil were analyzed to generate insights on factors influencing attendance. Variables such as patient's age, neighborhood, if patient received an SMS reminder, or if patient is enrolled in the Brasilian welfare program Bolsa Família, were all analyzed to find patterns and trends.
The UEFA Champions League WikiPedia page was scraped using the BeautifulSoup library and the 'read_html' function of the Pandas library. Details about players and clubs were imported into a DataFrame for analysis with Python.
A brewery database was queried and analyzed using PostgreSQL to figure out high performing territories, profit generating countries, top performing brands and salesperson, and low reach regions. Data Query Language (DQL) was used extensively, coupled with advanced SQL queries.
Stakeholders are mostly concerned with insights, and it's been proven that insights are better understood with visuals than with numbers. For this reason, I have studied extensively on the appropriate chart types suitable for a variety of audience and the perculiarity of the insight to be communicated. The link below contain dashboards ranging from Medical and Health data, to Retail and Sales data.