Intel® Analytics Toolkit
Intel Analytics Toolkit
Intel Analytics Toolkit is currently in a Private Beta. Contact us to particpate.
The revolution in big data is transforming industries and research, while spawning new solutions to a range of societal challenges. Big data strategies usually begin by capturing high volumes of varied data using Apache Hadoop*-based platforms that have massive scalability, cost effectiveness, and a vibrant open source ecosystem. But once the data is captured, achieving anticipated insights remains elusive. Data science expertise is in high demand but hard to find, a problem that is exacerbated by the added skills needed to program across a myriad of open source tools, and by workflows that are inefficient for iteration and collaboration. Finally, tools used are often geared to answering known questions, with limited workable methods to easily find hidden signals in data patterns and connections.
The Intel® Analytics Toolkit addresses these barriers to achieving value from big data and enables data scientists to achieve greater insights more quickly and with reduced complexity. A simpler programming environment lets data scientists focus on analytics instead of mastering the details of programming to Hadoop and the myriad of open source tools. Data scientists can orchestrate and easily iterate through the end-to-end analytics workflow in a single program, using a familiar programming language that executes analytics using fully scalable algorithms. Out of the box, the platform unifies entity-based machine learning with an end-to-end graph processing pipeline that includes powerful algorithms for uncovering relationships hidden in big data. The modular framework enables users or developers to extend and integrate new analytics functionality and algorithms.
By bringing simpler analytics programming and the full range of graph processing capabilities to the Hadoop “data lake," the Intel Analytics Toolkit is helping to democratize and accelerate big data powered solutions.