TDWI Checklist Report Helps Enterprises Explore Big Data Analytics

SEATTLE, Wash., March 13 — TDWI Research has released its newest Checklist Report, Utilizing Big Data Analytics with Hadoop. The report examines how enterprises can gain competitive advantage using advanced analytics and how several technologies are coming together to form the fabric of an analytics ecosystem.

Increasingly, companies are dealing with larger amounts of diverse data, some generated by newer sources (such as smartphones) and much of it unstructured (such as machine-generated data). In addition, as big data gets bigger, companies are looking at technologies that can handle the deluge, including Hadoop, an inexpensive solution for storing and processing big data. “[Hadoop] is rapidly becoming an important part of the big data ecosystem,” writes the report’s author, Fern Halper.

“Advances in analytics algorithms and analytics processing have also helped organizations cope. Visualization has helped companies explore data to discover insights—even with big data,” Halper explains. “Analytics algorithms such as machine learning and predictive analytics have matured to support the distributed processing needed for big data analytics.” Text analytics is also helping enterprises derive new meaning from unstructured data.

Halper begins the report by discussing the fundamentals of Hadoop and provides an overview of the collection of tools that exploit its power. She also explains key technologies that accelerate processing and return answers more quickly to business users, such as in-memory analytics.

The Checklist Report explores the role of ETL (extract, transform, and load) and data preparation to enable big data analysis, techniques for enhancing big data exploration and insight discovery (such as visualization techniques and descriptive statistics), and advances in analytics (including text analytics and other data mining techniques).

“Our report helps enterprises understand how text data fits into the analytics mix—including e-mail messages, call center notes, tweets, and blogs,” says Halper. “Much of the data in a typical Hadoop cluster is text data, so enterprises will want to take advantage of it to gain a more complete picture of what is happening with their customers and with operations.” Of course, these new technologies will take a new skill set; Halper looks at what skills data scientists will need to master to take advantage of big data.