Amit Kayal

Recently Published

Transport Mode Analysis with SVM Mode
The attached data has records of 444 employees in a firm. The variables are described below: -Build a model that best explains the employee's decision to use cars as the main means of transport? What would your predictions regarding their choice of transport be for the following two employees? The model explains how response variable with multi level values is being dealt with SVM and multinon regression
Employee Retaintion Analysis
This solution explains how logistic regression and further regularization is used to develop a simple model for predicting employee retention.
Time Series Analysis of Sales
This solution explains how HoltWinter method and ARIMA mode can be used for prediction
In this solution, I am going to build a Knn classifier using R programming language. Will use the R machine learning caret package to build our Knn classifier. The most commonly used distance measure is Euclidean distance. The Euclidean distance is also known as simply distance. The usage of Euclidean distance measure is highly recommended when data is dense or continuous. Euclidean distance is the best proximity measure.
Naive Bayes Modelling
This solution explains how prior probability information can be used to predict new class.
Naive Bayes Algorithim and Predicting the class
This solution explains how Naive Bayes algo can be used for predicting categorical variable class
Classification and Prediction using KNN
This example takes data of cancer and uses KNN algorithm to predict the outcome. Solution also explains how we can find out optimal value of K which is required for predicting accuracy
Brand Analysis with logistic Regression
Solution explains how we proceed with logistic analysis and calculate mcfaden R2..
Classification Based Solution with KNN
This solution explains how LDA/KNN can be efficiently used for classification and prediction. Note: Here most of the predictor variables are categorical.
Marketing Response Analysis through Discriminant Analysis
This method explains how discriminant score can be applied to calculate the binary response variable
conjoint analysis
This simple snippet shows how conjoint analysis can be used to measure the importance of feature and level
Machine Learning Algorithms in R (random forest case study)
This case study explains how random forest model can be used for marketing data analysis and process of model fine-tuning
Prediction using logistic Regression Model
This script shows how logic regression can be used for prediction and its model evaluation methods
Analysis of Two sample and understanding of their difference
This analysis shows how we can compare two different sample and conclude whether they are having similar mean or not by using insurance sales data of two separate insurance..
Loan Approval Prediction Model
Company wants to automate the loan eligibility process (real time) based on customer detail provided while filling online application form. These details are Gender, Marital Status, Education, Number of Dependents, Income, Loan Amount, Credit History and others. To automate this process, they have given a problem to identify the customers segments, those are eligible for loan amount so that they can specifically target these customers. Here they have provided a partial data set. Problem link is
Decision Tree Model Evaluation
This examples shows how the decision tree helps us to identify important variables from the dataset.
K Means Clustering Analysis
This analysis uses a Bollywood music data and explains how clustering can be applied to separate them for further analysis
Independent Variable and their impact on Regressor
This project explains how we can understand the impact of independent variables on regresor or response variable
Regrssion Analysis with Variable Transformation
This analysis shows regression analysis step by step. Also explains how we can include variable transformation for better prediction.
Analysis of Cereal Survery through Factor Analysis and Bramd Interpretation
This analysis shows the power of FCA and a process of Identification of latent factors from variables. FCA allows us to reduce number of variables through latent factors identification and then we can use for dependent variable interpretation.
Response variable skewness and decision tree impact
This analysis tries to explain the issue we face in CART when response variable is skewed...