People Searches Made Easier With Data Mining

A new personal data mining technology could make people searches and background checks faster and more reliable.

Many companies that collect people’s personal information have large amounts of information on individuals, but don’t have good technology for searching and analyzing all of this personal data.

Suresh Venkatasubramanian, a computer scientist from the University of Utah, has developed a data mining tool that makes the searching and analyzing of this personal data faster and easier.

Search engines like Google; Social networks like Facebook and retailers like Walmart as well as government departments are in the process of creating profiles of people. These people profiles contain hundreds of details on a person.

Your web searches, retail purchases, Facebook likes and social network of friends are all stored in multiple corporate and government databases.

Each of these details say something about a person. The trick is mining all of these details to get an overall picture of someone.

Venkatasubramanian and his coworkers have created a new way to mine multidimensional data on people.

He says that previous data mining methods have trouble analyzing personal information from more than 5,000 people. The new data mining method can easily crunch numerous details on more than 50,000 people.

Analyzing numerous details about people poses a problem because each personal attribute can impact other details.

Some common examples of companies using personal data mining include Amazon’s product recommendations to buyers based on their past purchases as well as the purchases of people with similar interests.

Netflix uses a similar personal data tool for recommending movies and Facebook recommends friends based on people who already are your friends as well as friends of friends.

The ever growing volume of this personal information from search engines, social networks and giant retailers is what makes the mining of details on people so difficult.

The new data mining technology can handle large amounts of personal data because it analyzes people’s attributes incrementally, rather than all at once. This speeds up the data mining of people because you can start without having all the information and mine a person’s information as you go.

Venkatasubramanian will discuss his new personal data mining tool on July 28 in Washington at the Conference on Knowledge Discovery and Data Mining, which is sponsored by the Association for Computing Machinery.