jisilvia Goto Github PK

basic_regex

A regular expression is a sequence of characters that define a search pattern. Usually such patterns are used by string-searching algorithms for "find" or "find and replace" operations on strings, or for input validation.

covid_19_reopening

Given its current circumstances, should a county reopen to prevent a second outbreak of COVID 19? We will answer this question by identifying similar counties, which can be referred to when deciding whether a county should expect a spike in infection rates if it were to reopen given the current circumstances.

decision_tree_algorithms

This project aims to build various Decision Tree Classifier Models to predict the income group of people along with seven demographic variables. Out of all decision trees, the best model is chosen, refined, and visualized. Subsequently, the model is automated and predictions are made for a new record.

eda_indian_air_pollution

This project analyzes levels of various pollutants in India by state. The results are used to calculate air quality, which is then compared among the states with a focus on Andhra Pradesh.

ensemble_models

In this project, four predictive machine learning models are applied to one dataset in order to select the model that returns the most accurate predictions. Different values for the hyperparameters were used in order to optimize model performance by obtaining the highest accuracy and AUC scores.

knn_recommender_system

The objective of this project is to build a kNN-based recommender system in order to predict the top 5 movie based on a given movie, in this case "The Post". As there is no need for classification or regression, the nearestNeighbors model and neighrbors() method are used to find the 5 most closely related films.

logistic_regression_heart_disease

This project employs a Logistic Regression model with the objective to predict the risk of Cardiovascular Disease and identify factors that increase may increase risks.

naive_bayes_spam_mail_detector

The model is trained with a set of emails labelled as either from Spam or Not Spam. There are 702 emails equally divided into spam and non spam category. Next, we shall test the model on 260 emails. The model is to predict the category of this emails and compare the accuracy with correct classification that we already know.

qdoba_survival_analysis

sql_data_requests

This project demonstrates an understanding of various SQL syntaxes to complete multiple data requests, including custom requests, using the Sales_Order and sakila databases.

sql_grouping_subqueries_joins

This project demonstrates experience summarizing Subqueries and Joins in SQL to fulfill multiple data requests.

sql_indexes

This project uses the a custom database table in combination with various Index functions to fulfill multiple data requests.

sql_joins

This project uses the sakila sample database which contains customer- and inventory information for a movie rental business. Inner Join and Left Join statements are used to execute specific tasks such as creating email lists, checking inventory, and identifying movie genres.

sql_string_date_functions

This project demonstrates an understanding of Date and String functions by fulfilling various data requests. It also demonstrates knowledge of writing and executing SQL within Jupyter notebook, which is traditionally used for Python code.

sql_union_case

This project demonstrates experience using UNION and CASE functions in SQL to fulfill multiple data requests.

sql_windows_and_cte

This project uses the sakila database in combination with various window functions and CTEs to fulfill multiple data requests.

wine_quality_analysis

Several wines are analyzed in order to gain insights on what features affect a wine's rating using logistic regressions models, kNN algorithms, topic modeling, LDA, and Gensim models.