In this age of heightened vigilance, the sheer volume of information collected by law enforcement agencies might seem overwhelming. According to the National Criminal History Improvement Program (NCHIP), since 1995 the number of criminal history records held nationwide grew by 29 percent while automated records increased 35 percent. Over the same period, the number of records available for sharing under the FBI’s Interstate Identification Index (III) climbed 75 percent.

But collecting and archiving information is just the tip of the iceberg. The real challenge is to quickly locate specific individuals using attributes such as name, date of birth, address, social security number and other data despite any attempts at deception. This can be difficult even when no attempt is made to hide one’s identity. Anyone who has ever searched a telephone directory for a name like RICKY SMITH only to encounter variations such as RICK SMYTHE, RICHARD SMITHE, RICHARD A. SMITH, RICARDO A. SMITH and R. A. SMITHE realizes the demands such a search entails.

With the continuing threat of international terrorism, the ability to distinguish one foreign name from another also poses unique challenges. Anglicisation, when FRANCO becomes FRANK or JORGE becomes GEORGE, is a potential source of confusion. Another, for example, is the inability of many computer search programs to discern that EL ABIR ABAHAR, ABAHAR EL ABIR and MHMD ABIR ALABAHAR are all variations of the same Middle Eastern name.

One organization facing these challenges is the Florida Department of Law Enforcement (FDLE) whose Office of Information Resource Management provides the technology needed to maintain and share information throughout the state’s law enforcement community. The FDLE’s core system is the Florida Crime Information Center II (FCIC) which is available to more than 1,000 agencies and 52,000 stations. These include sheriffs’ offices, police departments, jails, corrections and probation agencies, state and national criminal justice agencies, the Florida Legislature and average citizens via the group’s Web site (http://pas.fdle.state.fl.us). In all, the system processes more than 1.8 million transactions daily.

Researchers and investigators at FDLE and other criminal justice organizations are often under intense pressure to make timely and difficult decisions based on search results that include supplemental and fraudulent names and social security numbers. False “hits” and information overload make finding the most accurate and useful information a lengthy process when time is always in short supply.

Brenda Owens, FDLE’s Chief Information Officer, aptly illustrated the elusiveness of lawbreakers and the difficulties inherent in identifying them by describing an individual from the state’s wanted list. “This person had 42 aliases, 14 dates of birth and five social security numbers,” she said.

Deliberate deception along with innocent errors and variations in data are unavoidable problems that plague the information hub of every law enforcement organization. Because these databases typically collect information from a variety of sources, with differing data structures and data verification processes, it is inevitable that false, duplicate and inconsistent records will make their way into the system.

To counter these deceptions and other identification-related problems, FDLE has implemented sophisticated identity search and matching technology from Search Software America (SSA) of Old Greenwich, Conn. SSA develops and markets software products that significantly enhance an organization's ability to search, find, match and group identity data within computer systems and network databases.

Computer systems that search and match identity data must penetrate a fog of spelling and typing errors, transcription mistakes, missing or extra words, initials, word sequence variations, concatenated names, abbreviations, foreign names, unpredictable transliteration from non-Latin alphabets and much more. This is compounded by the volume of information records being searched and the need to do it all in real-time.

“On a daily basis the FCIC II system, with the SSA software, catches multiple wanted, missing and runaway persons---even when there are name permutations and altered dates of birth,” said Jerilyn Ammons, Data Processing Manager for the FCIC II system. “We have had successes ranging from the U.S. Coast Guard apprehending a wanted boat crew member on boarding to officers running a juvenile’s name to find that he has been reported as a runaway. Timely and accurate responses from FCIC II are critical as officer and public safety are at stake. Efficient and intelligent name searching capabilities are very important to the system.”

Before implementing SSA software the FCIC system used an in-house system based on Soundex, which classified names into groups whose surnames were more or less phonetically similar to each other. However, phonetic search techniques like Soundex and NYSIIS alone can’t address all aspects of the problem and are not able to efficiently process data from multiple countries, character sets or languages.

The safety of Police Officers and the Public is paramount to the Law Enforcement community. The need for reliable and accurate identity search results must not be underestimated or overlooked in this continuing challenge. And of course, such accuracy can also mean lower inconvenience for the law-abiding public.

(For more information on SSA’s identity search and matching software products, visit www.searchsoftware.com).