Development Of An Information Retrieval System Using Tree-structured Clustering
[FRSC Benue State]

1.3    Justification for the Study
This study provides a means of easy storage and retrieval of information of vehicles and their owners for the FRSC in Benue State. It eases the stress of searching through the entire directory when retrieving information on an existing record; it will ensure the provision of a clear statistics of vehicle owners in a particular local government in the state. The output of the study shall serve as a benchmark for the Federal Road Safety Corps on the application of tree structured clustering in information retrieval and the study will also serve as a reference material to those who use this project material.
1.4    Aim and Objectives of the Study
This study is designed to help Federal road safety corps in Benue state for easy registration of vehicles and their owners, and efficient retrieval of this information anytime and anywhere in and out of the state. The following are the objectives of the study:
i    The review of clustering technique as a method of structuring data for easy storage and retrieval as an alternative way to the manual way for storing and accessing the information.
ii    To develop a system that will minimize or curb the fake registration menace in Benue state.
iii    To evaluate the system performance based on, accuracy, speed and safety.
1.5    Scope of the Study
The system thus developed is based on the retrieval of records of vehicle owners registered to different local government area. A case study of Benue State, this system can be used by a government agency (Federal Road Safety Commission) majorly for quick and easy access of registered records, through it; records can be easily updated or modified.
1.6    Definition of Terms
The terms used during this project work are as defined thus:
Cluster: a logical amount of disk space that can be allocated to hold a file or directory. Algorithm: a sequence of steps that is used to find solution to a particular problem Database: a computerized or automated record keeping system
Query: formal statements of information needs.
Oracle Database: a relational database which is queried using a Structured Query Language.
Hierarchical clustering: a method of data analysis which seeks to build a hierarchy of clusters.
FRSC: Federal Road Safety Commission. A government agency charged with the statutory responsibility of road safety administration in Nigeria.
Information Retrieval (IR): a discipline involved with the organization, structuring, analysis, storage, searching and dissemination of information.
Cluster Centroid: the point with coordinates equal to the average values of the variables for the observations in that cluster.
Multivariate datasets: a collection of data items that contains large and multiple variables.
Dendogram: a tree diagram used to illustrate the arrangement of clusters produced using hierarchical clustering.
Categorical Data: values or observations that can be sorted into groups or category.
Numerical Data: values or observations that can be measured.
Number plate: a metal or plastic plate attached to a vehicle for official registration and identification purpose.