Computer-Based Transaction Processing
Computer-based TPSs are often considered the “bread and butter” of the management information system application. No matter how nervous upper management in a medium to large organization is about spending in the information system area, it knows that it cannot pull the plug on its TPS and survive. Actually, many large companies have had computer-based TPSs since the 1950s. Most TPSs have been—and still are—mainframe oriented. IBM equipment and their compatibles currently claim the lion’s share of the transaction processing marketplace (Mahar, 2003;Tessler, 2015).
Although many companies consider the TPS to be their most important computer application, a surprisingly large number of firms have not carried computer-based information processing far beyond the transaction processing stage. TPSs in many organizations today are used in this way as competitive weapons. Additionally, the move from dumb terminals to intelligent microprocessor-based workstations is expected to alter transaction processing in other ways, such as by distributing certain traditionally mainframe-based centralized transaction processing functions closer to relevant offices or departments.
As indicated, the TPS supports the processing of an organization’s many transactions.This includes accounting for the transactions on its records, as well as providing support activities such as sending out payment reminders. Recently, gaining competitive advantage has become a TPS concern in some firms, especially those that are working to tie customers and suppliers more closely together with the organization’s TPS via electronic linkages.
Role of Information Technology and Transaction Processing
For many businesses, a transaction refers to an exchange of goods or services for money. The earliest TPSs were manual systems. A clerk would record transactions in a journal or on numbered, multipart forms. These transactions would later be transferred manually to a central system of handwritten records or file folders responding to individual customers or suppliers. These records would be set up to trigger statements to customers or checks to suppliers. Many small businesses still operate with manual TPSs;
however, inexpensive and easy-to-use computer technology' is finding its way into more small businesses (Mahar, 2003).
On the other hand, and at the same time, IT has been developing analytical systems. One strategy for extracting meaning from large amounts of investigative information is the use of data mining applications. Data mining systematically searches information to identify relationships and patterns. Although data mining has been used effectively in private industry for a number of years, law enforcement has trailed behind in the application of this technology. As an interesting comparison, data mining techniques in the commercial environment have allowed retailers to know more about purchasing habits than the police know about criminal suspects.
To explore the use of data mining within criminal justice, particularly criminal investigations, data mining software analyzes relationships and patterns in stored transaction data. Data mining software consists of sophisticated search programs, advanced statistical techniques, and innovative graphics features. Search programs used in data mining software provide users with the ability to make queries that use varied search criteria and repeatedly redefine those criteria to make searches as useful as possible. By using data mining software, investigators can initiate database searches that extract information describing relationships between persons, events, and other aspects of criminal activities. Data mining systems provide users with graphic displays that make it easier to see the detected relationships or patterns.
A typical law enforcement data mining application might attempt to identify a suspect when the only available information is a crime report and a vehicle description. An investigator could initiate a query of a regional network database to obtain information that would identify a suspect. Data mining software would then search information compiled by all agencies participating in the network. The vehicle description contained in the crime report submitted by one agency might match an entry in a field interview report submitted by a different agency.The field interview report might indicate that the vehicle was seen a short distance from the crime scene at a time close to the time of the crime, and that its driver had been questioned and provided a name and address. Data mining software could then be used to determine the involvement of the now-identified suspect in other crimes. Without the advantage of data mining software, information from the crime report and the field interview report might never be found or linked.
The 2002 “DC sniper” (or “beltway sniper”) case investigation illustrates the difficulty in searching massive amounts of information available to law enforcement agencies. Because the various shootings in this case took place in Washington, DC, Maryland, and Virginia, multiple law enforcement agencies were compiling information, resulting in the availability of a large amount of data in various systems. Review of the investigation revealed that information on the vehicle used by the snipers had been previously reported by law enforcement agencies, but the volume of data and their storage in disparate systems precluded timely searches. Data mining addresses this kind of problem.