GraphLab Create.

A machine learning modeling tool for developers and data scientists.

Application-Oriented Toolkits


Graphlab Create™ simplifies development of machine learning models. Each incorporates automatic feature engineering, model selection, and machine learning visualizations specific to the application.



Recommender

Recommend products, ads or services to increase revenue and engagement.

View More
Data Matching

Identify and link records within or across data sources that correspond to the same real-world entities.

View More
Sentiment Analysis

Score & summarize sentiment on text corpora to efficiently understand issues in support forums, social media, & reviews.

View More
Churn Predictor

Predict the likelihood customers will churn, understand the influential factors, and take action to prevent it from happening.

View More
Lead Scoring

Predict which trial users & prospects are most likely to convert to customers to prioritize your sales efforts.

View More

Elements for Advanced Machine Learning Applications


Common data structures, a broad set of algorithms, and components that are open, scalable, and extensible empower you to design advanced and innovative machine learning applications unique to your company and data.

Scalable Data Structures
manipulate & transform tables, graphs, & time series
Feature Engineering
compose & share data pipelines using stateful transformers
Anomaly Detection
local outlier factor, moving z-score, & bayesian changepoint detection
Classification
random forest, boosted trees, SVMs, neural nets, nearest neighbors
Deep Learning
create RNNs & CNNs, build with mxnet primitives, or choose from our models
Nearest Neighbors
standard & composite distances, ball-tree & LSH
Text Analytics
topic modeling, tfidf & bm25, tokenization, & inverted-index search
Pattern Mining
top K FP growth for mining of frequent item sets
Image Analytics
transform images for tagging, search, and feature extraction
Regression
linear regression, decision trees, random forest, boosted trees
Clustering
dbscan, kmeans
Graph Analytics
label propagation, pagerank, connected components, shortest path, and others
Model Optimization
find the best model with model search, evaluation & cross validation
Machine Learning Visualizations
build trust with interactive evaluations, explanations, & explorations
C++ SDK Plugin Architecture
easily extend Turi toolkits with your own