Академический Документы
Профессиональный Документы
Культура Документы
Contents:
1. Introduction ...................................................................................
3. DI Technology ..................................................................................
4. Architecture .....................................................................................
5. Features ...........................................................................................
6. Advantages ......................................................................................
1. Introduction
“The world is becoming digital by the day – for good as well as bad”
We live in an age where most of the information is raw digital data. The
information may consist of images from a satellite orbiting our planet, a credit
card transaction verification system or our criminal or interrogation reports.
However the problem with all kinds of raw data remains the same - What to do
with all this data?
The value of raw data is typically predicated on the ability to extract higher level
information: information useful for further investigations, decision support, for
better understanding of the crimes plaguing a state, for storage of all data in a
structured format.
“Day by day more and more data is being piled up so what should we do with this
data? The answer lies with the problem itself. Since we have this data we can use
it to its own advantage by using data mining and extracting useful information it”
The process of Data Mining – To analyze data and find hidden patterns in it has
been in vogue since centuries. However, with the rapid increase in the inflow of
unstructured digital data, sophisticated statistical analysis and modeling
techniques are now used to generate patterns and relationships hidden in raw
data.
Today, multiple Law Enforcement agencies across the world are using statistical
analysis and pattern matching to catch suspects and criminals or just to draw
intelligence out of the humongous quantity of raw data available with us.
2. The Problem
“Criminals leave their traces behind huge amount of data – however manual
analysis of this data usually takes more time than it took to gather the data”
As the focus on extracting intelligence from raw data has sharpened, Law
enforcement agencies need information that allows for quick action at all levels.
However, most of our agencies have fragmented systems and maintain
information in disparate systems, thereby making it difficult to find and access
the right piece of information quickly. Reports can take days or weeks to produce.
Answers to queries vary, based on which system is used.
The problems –
Data Analysis – The rapidly increasing inbound flow of raw data has created
the need for automated extraction of useful knowledge from huge
amounts of data which not only helps in finding new information about the
criminals, their associates or modus operandi etc. but also helps in better
understanding of them and their activities
3. DI Technology
For any country the biggest problem with digital data is how to create it, how to
store it and how to analyze the data. The typical police department has many
operational systems that capture and report crime-based information. The
challenge is integrating data from fragmented systems to provide a single view,
finding the ability to analyze crime trends using data from all systems,
accurately reporting crime statistics, and complying with state and federal
reporting requirements.
3.1Data Warehousing
3.2Data Mining
The value of raw data is typically predicated on the ability to extract higher
level information which leads to the need for automated extraction of
useful knowledge from huge amounts of data. Knowledge discovery and
data mining are techniques to discover strategic information hidden in very
large databases
Algorithms used
o Distance Based
Simple
K nearest neighbors
o Neural Network Based
Propagation
NN supervised Learning
o Rule Based
Generating rules from Decision tree
Generating rules from Neural network
o Association Techniques
Sampling
o Partition algorithms
Minimum spanning tree
PAM algorithms
Logical analysis is done using Data and Task parallelism algorithms The application
is programmed to perform investigative analysis on the data stored in data
warehouse. The application will be setup to automatically identify the data from
the input media and alert the administrator about the input option. The
application also maintains an index of each database and correlates the
information so that the search takes place faster and easier.
4. Architecture
4.1Hardware Architecture
5. Features
Software designed using state police departments requirements in
view
Flexible thresholding
User friendly GUI Interface
High level of security (Multifactor Authentication)
Possibility of connectivity between district, state and police station
databases
Reduces repetitive manual work
System helps officers –
o Draw Patterns and analyze a given situation graphically
o Draw links and associations between multiple entities
o Link past crimes of an individual
o Analyze the progress on a particular case
o Study and analyze crime trends in respective jurisdictions
6. Advantages
The main objective behind developing and completing this product was to
help our Law Enforcement agencies save their energy and the time which
goes into manual analysis of the humongous data which most of the time
does not lead to fruitful results. The product not only does the analysis but
helps in creating a database which serves the objective. The automated
tool leads to discovery, extraction and linkages of sparse evidence
contained in large amount of classified and unclassified data resources. It
has detection capabilities to extract relevant data and relationships about
people, organizations, and activities from source data. It will link items
relating potential criminals, terrorist groups or scenarios, and learn
7. Case Study
“The world is already using the best technology available to combat crime,
terrorism and other facilities to make their tasks easier. Why should we be left
behind?”
Snapshot by agencies during the 9/11 investigation where similar tools were
used to discover the relationships between various terrorists