Facilitating Trade:
Improving Customs Risk Management Systems
In the OIC Member States
42
Figure 11: Data Mining Prediction Concept
Source: Author’s Compilation
CRM and intelligence must evaluate past predictions and actions captured. The feedback loop
lets the predictive models grow smarter and helps CRM to focus effort in the areas.
Data Mining - Crime prediction and prevention analytics from data mining can assist OIC MS
CRM to make the best use of the resources and information and to measure and predict crime
and crime trends. Mining of the LE data provides insight that lets CRM and intelligence to track
criminal activities, predict the probability of crime/incidents, effectively deploy resources and
solve cases faster.
The data mining can assist the CRM in following perspectives:
 Instrumented - Information/enforcement records collected from multiple data sources
and analyzed for hidden patterns and relationships that are vital to fighting violation of
Law;
 Interconnected – DW, BI and data mining can provide CRM with quick and reliable
access to easily understand analytic crime forecasts based on historical data,
intelligence, open sources, etc.;
 Intelligent - Criminal behavior, patterns, and proactive tactical enforcement decisions -
generated in predefined time frame or ad – hoc basis, on the dashboard, reports, and
analysis CRM will need to extend the domain of the data, by implementing the text
mining techniques that are capable of extracting knowledge from text data about
something that was previously unknown.
Text Mining - Text databases are rapidly growing due to the increasing amount of information
available in electronic form. This includes electronic publications, news articles, research
papers, books, digital libraries, e-mail, etc. The Worldwide Web can be used as an
interconnected, dynamic text database. The data and information should be stored in the form
of structured text databases. Unlike the field of database systems, focused on query and
transaction processing of structured data, in text mining is a way to organize and retrieval of
information from a large number of text-based documents. The goal of text mining is to discover
or derive new information from data, finding patterns across datasets, and/or separating signal
from noise (or snowflakes). There are many approaches to text mining, which can be classified
from different perspectives. The approaches are differed on the inputs in the text mining system