.jpg){kind=link}
Exploring my data science portfolio
This is a simple classification problem trained on python scikit-learn library.The classification model takes the independent variable eg. age,sex,cholestrol,blood pressure etc.,from heart attack data set to predict whether the person will get heart attack or not.
In this project chi-squre test has been used for to checkfeature importance of categorical variableand independent t-test is used to compare the mean of variables grouped on the basis of output category and by finding correlation among numerical variables to get the importance of each variable in deciding output.Features renaming also done with help of google,you can find here.
We also use Pipeline method to apply 10 classifiction algorithms(1.logistic Regression 2.Decisiontreeclassifier 3.Randomforestclassifier 4.GaussianNB 5.KNN 6.Gboost Classifier 7.AdaboostClassifier8.SGDClassifier 9.SVC 10.MLP Classifier) to get the best accuracy which i have got in KNN=82.4%.Then i also apply for loop to get the best random state producing good accuracy and then we have got accuracy of 90% with choosing appropriate n_nieghour parameter which is n_neighour=6. After plotting AUC-ROC curve we have got the value AUC=93% which is a good value for a model.
.jpg)
This is a regression based ML plroject build on python scikit-learn library.The aim of this project is to predict the price of the car on the basis of its given independent feature eg. model, engine feul type, engine HP, engine cylinder, transmission type, driven_Wheels, number of door, market category, vehicle size, vehicle style, highway MPG, city mpg, Popularity.
We applied 11 ML regression algorithms to build our model, then we get best accuracy in Extra tree regressor which is 97%. After analysing the features and correlation we found 'Engine HP' is the most important factor for the price of the car.
This is a Regression problem trained on python scikit-learn libarary.The data is related to civil engineering/Architecture where compressive strengh of material being used is import factor to determone the stability, sustainibility of building/bridge/construction.The target aim of the model is to predict the compressive strength of concrete on the basis of independent variable -cement,Blast Furnace Slag,Fly Ash,Water,Superplasticizer,Coarse Aggregate,Fine Aggregate.
we used 12 regression algorithms to build our ML model where we have got the best result with Extra Tree Regressor having accuracy of 97%,RSME=4.08 which is very good for a model.
this is a unsupervised machine learning problem where diifferent segment or group of customer has to be made using K-mean clusterring and hierarchical clustering algorithm . After taht we have to identify our target customer who has more potential to give profit. You can visit the data by clicking on the image below.
This is visual analysis based project where we have the data set of 45 countries. Using plotlly chroropleth, seaborn, matplotlib we made different plot to analyse the count of daeth, rate of death due to covid-19 in different countries with respect to time. We also found that Mexico has the highest count and rate of death.For more detail about the dataset please click on the image below
This is a imbalance classification based ML project where we are given a direct marketing campaign (phone call) dataset obtained from a bank. The data contains 3090 rows and each row refers to an individual customer record. The dataset contains 15 input attributes and one target or class label attribute (subscribed to term deposit).
We used 8 classification algorithms to build our model on sklearn. Smote technique has been used to tackle the imbalance of the data using imblearn package. We got the best accuracy with Decision tree model accuracy =92% and f1-score =.93, after analysing the correlation and feature importance we got the'duration' is the most important factor to decide whether customer takes subscription or not.
Banking dataset that helps running a campaign. how do banks select target customers so that chances of selling their product maximize, this will help you understand these things.The targeted customer has to be predicted on the basis of some features eg.age,education,marital status,salary,housing etc. We have build a classification model to classify the targeted customer. you can get the dataset from kaggle by clicking on below image
predicting the churn, we can react in time and try to keep the client who wants to leave. Based on the data about the services that the client uses, we can make him a special offer, trying to change his decision to leave the operator. This will make the task of retention easier to implement than the task of attracting new users, about which we do not know anything yet. We build the classification model to predict whether the customer will churn or nor.Here we have focused on recall rather than accuracy and we get recall upto 80%. You can find the dataset on kaggle by clicking on below image.
This is a linear regression based problem where the price of the house is predicted on the basis of its feature eg. no of bedroom, sqft area,floor waterfront,condition, age etc. we also find the main factor which determine the price of house. You can find the dataset on kaggle by clicking on below image
• Using Excel , the output of the topics modelling which has been applied on the data in python is converted into csv format and then imported into MySql. • Transformed the data using DAX calculation in Power BI and prepared the dashboard to extract the business insights in the data.
• I was assisted in developing Data warehouse using MY SQL to convert unstructured data into Fact and Dimension table for more informative and accessible data. • Prepared BI dashboard using Power BI for day wise, month wise ,year wise , hour wise mall entry analysis report