A causal analysis goes beyond simple associations between variables to estimate the causal relationships with a goal of answering whether a change in treatment variables (e.g. job training program) actually causes changes to outcome variables (trainees' real life earnings).
Counterfactual analysis allows you to estimate the impact of an intervention on prediction results. For example, answer what would sales be if we increase a product price? This works in our predictive analytics (Regression & Classification) by estimating the causal effect of the intervention on the prediction outcome using Causal AI.
De-biased predictions help you to predict unknown values or labels using known examples. Our predictions (Regression & Classification) are generated with a full explanation of which factors contribute to model decision and by how much. You can also easily remove certain associations or biases in data to improve prediction fairness.
Data Cleaning & Imputation helps you to detect empty or invalid fields and suggest fixes for those fields with Deep Neural Networks based on rules defined by users. User can also easily generate suggestions for missing values by simply selecting “Impute empty or null cells”.
Sentiment analysis can classify a text content into positive, negative, and neutral sentiments. Our sentiment analysis works at a context level and extracts the sentiment of a given text at sentence and entity levels. It can be used by businesses to better understand their customer’s perception of their brand, products, and reputation.
Our time-series forecasting with cutting-edge Deep Learning AI works seamlessly with missing values and multiple time-series forecasting. You can predict future values of variables based on their historical values. Users need only to select the date variable and prediction length.
Segmentation or clustering is an unsupervised machine learning technique with no defined dependent and independent variables. Our Deep Learning based Segmentation learns complex relationships between various characteristics from the unlabelled data to cluster the data into similar groups.
Correlation is a statistical association between variables whether causal or not but in general, correlation alone is not sufficient to infer a causal relationship. Correlation Analysis can be used to find if there is a linear relationship between variables and to estimate the strength of that relationship.