Você está na página 1de 3

UNIVERSITI UTARA MALAYSIA SQIT3033 KNOWLEDGE ACQUISITION IN DECISION MAKING GROUP A INDIVIDUAL ASSIGNMENT 1 PREPARED BY: CINDY LORETTA

MAXIMUS MATRIC NO: 212509 PREPARED FOR: DR IZWAN NIZAL BIN MOHD SHAHARANEE DUE DATE: 6 MARCH 2014

Discuss how data mining application can be utilised in construction projects All construction projects are constrained by schedules, budgets, safety as well as environmental regulations. From these constraints it makes the construction project management becomes more complex and difficult. Data mining is applied in the construction project as it can help to make the historical data to be used effectively and efficiently. In general, data mining process can be divided into 4 main parts which are aggregate the data set for analysis, put forward data to data mining tool, analyse then interpret result and lastly applying the result to solve problems and situations. From this data mining process, a result will be known which then can be used and applied in practice to handle and as well as monitor the risk especially in the construction project. Here is one example of data mining application used by the US Department of Agriculture (USDA) . Data mining defined as nontrivial process of identifying valid, novel, potentially useful and ultimately understandable patterns in data (Advances in Knowledge Discovery & Data Mining, U.M Fayyad et. al., eds., MIT PRESS, Cambridge, Mass, 1996) . It frequently uncovers the patterns that can predict future behaviour and proved useful as for example it helped the rural housing service becomes better understand and clarify loans problems. This rural housing service issue by the USDA help them better understand the loans, improved their management of lending program as well as reducing the problem loans incidence. This department wants data mining to find patterns that can distinguish borrower who repay from those who dont. When it comes for them to build a data models, they divide it into two kinds of models which are predictive and descriptive models. A predictive model is a model that predicts the value of a particular attribute. For example, the revenue of customer will generate next year and an insurance claims likelihood of being fraudulent. Besides, in this modelling technique a sample decision tree is used to classify the loan prospects by income and marital status. Others samples are neural network, nave bayes and k-opt neighbour. While descriptive model is a model that consists of 2 important model types: clustering and association. After the model are built, it will be tested to reveal on how well a model predicts the target variable which in this case is the loan classification.

In addition, in any construction project, some necessary resource should be involved like humans, machines, materials etc. The data also cannot be limited as a lot of data is needed to conclude the results. If data is limited, it will be quite difficult to conclude results which then will leads to the restriction of the application of data mining in construction projects.

REFERENCES Gerritsen, R. (1999). Assessing Loan Risks: A Data Mining Case Study. Retrieved March 1, 2014 from
https://www.google.com.my/#q=data+mining+housing+loan

Zhang, J., MA, T., & Shen, Q. (2002). Application of data warehouse and Data Mining in Construction Management. Retrieved March 1 2014, from http://epub.uni-weimar.de/opus4/frontdoor/index/index/year/2004/docId/124

Você também pode gostar