Model Development Studio

Perform data profiling operations on structured and unstructured data, such as one-hot encoding, stemming, lemmatization, missing value imputation, and count vectorizer, etc.

Use built-in machine learning (ML) and deep learning-based techniques for dimensionality reduction, including singular value decomposition (SVD), principal component analysis (PCA), and restricted Boltzmann machine (RBM)

Use multiple options to model, including graph, ML, deep learning, and natural language processing

Perform model averaging techniques—stacking and ensembling

Develop models on massive-scale datasets by utilizing the in-memory distributed computing-based processing

Perform rapid model experimentation, development, and evolution through the visually intuitive drag-and-drop interface

Configure each node and drop it on the canvas with others to build your own model pipeline

Create and define your own features, based on separate boolean and aggregate operations with comprehensive feature engineering

Use the coding interface or the visual workflow editor to create new data columns

Create segments on both numeric and textual data

Create user-defined rules and conditions for segment creation

Make use of both macro and micro-level segmentation

Select the best models based on several visual performance metric reports

Evaluate the model performance using the rich set of evaluation metrics

Use multiple modeling techniques on the same feature engineered data with multi-model experimentation and evaluation

Access and configure all the modeling parameters

Fetch a detailed ‘feature importance report,’ explaining the output