Motivated by 24 Ultimate Data Science Projects To Boost Your Knowledge and Skills (& can be accessed freely)
Libraries: scikit learn, pandas, seaborn, keras
Dataset: 3 Iris species with 50 samples each and 4 properties, i.e., Sepal Length, Sepal Width, Petal Lenth, Petal Width (in cm).
Problem: classify 3 Iris species
Classification techniques: Logistic Regression, Naive Bayes, kNN, SVM, Decision Tree, Boosted Tree, Random Forest, MLP
Dataset: custumers' details such as Gender, Marital Status, Education, Number of Dependents, Income, Loan Amount, Credit History and others.
Problem: predict loan eligibility (Y/N)
Classification techniques: Linear SVC, SVC, kNN, Random Forest
Problem:
Classification techniques:
Dataset: 2013 sales data for 1559 products across 10 stores in different cities with certain attributes of each product and store such as weight, maximum retail price, size of store and so on.
Problem: predict sales
Regression techniques: Linear Regression, Neural Network
Dataset: 506 cases with 14 attributes for each.
Problem: predict NOX (nitrous oxide level) and MEDV (median value of a home price)
Regression techniques: Linear Regression, ElasticNetCV, LassoCV
Problem:
Dataset: Height and weight records of 5000 men and 5000 women.
Problem: Predict Weight/Height
Regression techniques: Linear Regression
Dataset: 5820 evaluation scores provided by students from Gazi University in Ankara (Turkey), total of 28 course specific questions and additional 5 attributes.
Problem: Cluster
Clustering techniques: KMeans, MeanShift, BayesianGaussianMixture, AgglomerativeClustering
Every notebook attachs helpful reading-materials. Here are some general ones:
- Choosing models
- Choosing features