Browsing by Author "Arief, Akbar"

Advanced analytical practices such as data mining or predictive analytics are concepts that are increasingly vital in the area of large data sets. Voluminous data are collected over the years and it is important to assess the data quality for the value of information. Large amounts of data can contain knowledge in the form of patterns. What knowledge an organization especially the high-paced halal industry can get once data quality is assessed and data mining technique is applied? In this research, we have two objectives. Number one is to assess the data quality on the unstructured data collected from the Fourier-transform Infrared Spectroscopy (FTIR) instrument and number two is to identify patterns of halal and non-halal by applying decision tree technique for data mining. Cross-Industry Process for Data Mining (CRISP-DM) methodology is used in this research by adding data quality element on the preparation phase to stress out the importance of it before running a model for the dataset. RapidMiner is used to generate the prediction model by splitting the data collected into halal and non-halal substances and the result is visualized on its absorbance. Additionally, the performance vector analysis is also performed to make sure that the data is not over-fit. Among the result found in this study shows how big the difference of absorbance between the halal and the non-halal substance