Intelligent Inventory Miner Approach for Effective Inventory Forecasting and High Dimensional Inventory Data Management

A Venugopal, Jisha, M Scholar
2018 Journal of Network Communications and Emerging Technologies (JNCET) www.jncet.org   unpublished
Data mining is the most appropriate method for many applications. Inventory data mining is one of the most prominent applications in every business. Tracking and analyzing the market database is more complicated process. To overcome the inventory data management issues, an intelligent inventory mining technique IIMiner is proposed. These inventory data's are considered as time series data's with quantity and product brands. These time series datasets need more attention for business
more » ... process. There are several techniques were used in the existing researches. Those techniques are based on the statistical analysis with the historical inventory transactional data. Inventory data forecasting and detecting anomaly from the time series data is performed using the IIMiner framework. To achieve the effectiveness, the IIMiner proposed three phases, which initially performs the Frequent Inventory Item Mining (FIIM). The sales forecasting using Temporal Naïve forecasting model is proposed, which is used to predict the inventory values. And finally the inventory anomaly has been detected from the time series data using a semi-supervised classification method based on the support vector machines. Here the weighted batch based anomaly detection is performed. The new anomaly detection approach is named as weighted batch-support vector machines (WB-SVM). Finally the experiments and results proved the proposed system utilized maximum accuracy and effectiveness in inventory management and visualization.
fatcat:ycicnz5pmbgyheha6yhr3arok4