Machine LearningData WranglingVisualizationData Sources
Build, Predict, and Evaluate
Exploratory’s Machine Learning framework provides a simple and consistent experience for building, predicting, and evaluating the models from hundreds of the machine learning algorithms available in R.
Open and Extensible
With Exploratory’s open machine learning framework you can bring your own favorite algorithms and run them natively, which means you can build, predict, and evaluate the models with point and clicks, just like any other out-of-the-box machine learning models.
Linear Regression
Linear Regression is the most favorite tool of Data Scientists. If you want to predict numeric outcome this is your best friend who is simple yet reliable the most.
Logistic Regression
Logistic Regression is useful when you want to predict the binary outcome like TRUE or FALSE based on input data. For example, you can understand what makes your customers your loyal customers or what makes the students admitted to
Classification algorithm can be useful when you want to automatically categorize things like loyal customers, product quality status, fraud activities, spam emails, etc. based on the past data. You can choose from a wide range of classification models in R such as Decision Tree, Random Forest, Boosting Tree, SVM, etc.
K-means Clustering
By employing clustering techniques like K-means Clustering, you can let data find hidden characteristics or segments of customers, employees, locations, etc. so that you can target the right audience with tailored solutions.
Ensemble - Boosting / Bagging
Ensemble models such as Boosted Trees, Random Forest, etc. helps not only improve the prediction performance but also helps you find which attributes of the data have more impacts on the outcome you want to predict. You can quickly access to these state of art algorithms Data Scientists love without coding in Exploratory.
Survival Analysis - Survival Curve
Survival Curve is the gateway to understand your customers retention rates, customer lifetime value, employee attrition, product maintenance cycle, on the time horizon.
Survival Analysis - Cox Regression
Cox Regression helps you understand what makes your customer churn or become loyal customers, what contributes product failures a given time horizon, but also predict when the expected events to happen.
Multinomial Logistic Regression
Multinomial Logistic Regression is an extension of Logistic Regression, but unlike Logistic Regression, it can predict the target outcome that has more than two categories.
Time Series Forecasting
With the revolutionary forecasting algorithm called Prophet built by a team at Facebook, you can quickly forecast your time sensitive data within seconds. You can use it to forecast the next quarter revenue, next month page views, next year customer growth, and many more!
Anomaly Detection
Anomaly Detection helps you detect unusual activities like financial fraud, suspicious web site access, machine faults, etc. or evaluate the impacts of A / B Testing.
Sentiment Analysis
By understanding your customers sentiment at various social media outposts and community forums, you can engage with your customers better and cater right messages when needed. In Exploratory, you can score the sentiment either by word or by sentence.
Multi Dimensional Scaling
Multi Dimensional Scaling (MDS) helps you understand similarities of your customers, products, portfolios, locations, etc. in a visually intuitive way.
Correlation algorithms can help find a set of variables or measures that have similar movements. This would help you predict one variable if you know only another. This technique is often used before building Linear Regression models as part of the exploratory data analysis phase.