Shulou(Shulou.com)05/31 Report--

Analysts and IT managers said on Computerworld's BI that through predictive analysis of business operations and historical data, companies can gain a significant perspective on long-term benefit analysis.

Different from traditional business intelligence practices, predictive analysis focuses on helping enterprises collect operational intelligence based on historical data.

If applied properly, predictive analysis can enable companies to identify and respond to new opportunities more quickly, they say.

James Taylor, CEO of decision Management Solutions, said in his keynote speech that predictive analysis is especially useful when companies need to use a lot of data to make decisions quickly.

Taylor says predictive analysis practices can help companies minimize risk, identify fraud and seek new revenue opportunities in three key areas.

For example, predictive analysis can help companies fine-tune their ability to identify risks in areas such as loans and credit issuance or fraud in areas such as insurance claims, he said.

What is important, Taylor says, is that companies can better identify new revenue opportunities by embedding predictive analysis into operational data. For example, by looking at customers' historical purchase patterns, companies can make reasonable predictions about the types of promotions and coupons that may resonate with customers.

Blue Cross and Blue Shield (BCBS) is an organization that makes substantial profits from forecasting and analysis. As an institution that provides health insurance for nearly 1/3 Americans, the Blue Cross Blue Shield system has accumulated a large amount of data related to claims settlement over the years.

A few years ago, the BCBS Association and Blue Brand created a single database called Blue Health Intelligence (BHI) to integrate all claim information maintained by 39 companies belonging to BCBS. The database is one of the largest unidentified repositories of health care data in the world, containing information on more than 100 million claims.

BHI is an independent organization that provides a range of business intelligence services to provide better medical services for its members and changes the way BCBS manages it costs.

Swati Abbot, president and chief executive of BHI, said in a speech that BCBS, like other health insurance companies, needed to control rising costs, which contributed to this effort.

Abbott says a large portion of medical expenses are spent on care for people with chronic diseases. In fact, she says, the top 5% of health care users account for more than 55% of health care costs.

By applying predictive analysis to its vast treasure trove of claims data, BCBS is getting better not only at identifying risk factors for certain chronic diseases, but also at identifying people at high risk for such diseases, she said.

"We have a health score for every member of the health program," Abbott said, which represents the possibility that individuals need lifelong treatment for chronic diseases. BHI has even developed modules for specific diseases such as diabetes that can predict diabetes risk based on previous data, she said.

'our goal is to use this data to enable doctors to provide better and more targeted treatment for high-risk patients, thereby reducing their need for expensive long-term treatment, 'she said. Albert says the predictive model enables BCBS to shift to a more incentive-based health care model, in which doctors' performance will be motivated.

Online dating site Match.com is another company that relies heavily on predictive analytics to run its services. The company collects and maintains a large amount of information, some from subscribers and some from monitoring their interactions on Match.com.

Jim Jim Talbott, director of consumer insight at match, says the challenge for the company is to find a way to increase each user's revenue by providing the best possible match based on each user's preference

The task is complex, he says, because users may make a specific set of requirements for potential partners, but then interact with people who go beyond their specific preferences.

To meet this challenge, Match.com has developed a predictive model that matches people not only according to their stated preferences, but also based on their website behavior and interactions with others.

Taylor says companies interested in predictive modeling need to have a clear idea of their goals before they start. He says they need to know what kind of decisions their predictions and analyses will drive and work in reverse on that basis.

To develop a good forecasting model, he says, companies need to focus on defining a clear set of business rules for each decision, and then focus on driving the analysis of the best decisions.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.