No abstract

Abstract-Detection of pests in the paddy fields is a major challenge in the field of agriculture, therefore effective measures should be developed to fight the infestation while minimizing the use of pesticides. The techniques of image analysis are extensively applied to agricultural science, and it provides maximum protection to crops, which can ultimately lead to better crop management and production. Monitoring of pests infestation relies on manpower, however automatic monitoring has been advancing in order to minimize human efforts and errors. This study extends the implementation of different image processing techniques to detect and extract insect pests by establishing an automated detection and extraction system for estimating pest densities in paddy fields. Experiment results shows that the proposed system provides a simple, efficient and fast solution in detecting pests in the rice fields.Index Terms-Automated pests management, image analysis object detection, object extraction.

Abstract-This paper focuses on the enhanced initial centroids for the K-means algorithm. The original kmeans is using the random choice of init ial seeds which is a major limitation of the orig inal K-means algorith m because it produces less reliab le result of clustering the data. The enhanced method of the k-means algorith m includes the computation of the weighted mean to improve the centroids initializat ion. This paper shows the comparison between K-Means and the enhanced KMeans algorithm, and it proves that the new method of selecting initial seeds is better in terms of mathemat ical computation and reliability.

Abstract-This study focused on the generation of the licensure examination performance models implementing PART and JRip classifiers. Specifically, it identified the attributes that are significant to the response attribute; it generated prediction models using the PART and JRip classifiers of WEKA; and it determined how likely is a reviewee to pass the LET. The respondents were obtained from the Education graduates of Isabela State University Cabagan campus who took a LET review and eventually took the September 2013 LET. The results obtained indicate the significance of the mock board exam, general weighted average of the reviewees in GenEd and MajorCore in predicting LET performance. The reviewee is predicted to fail the LET if he will obtain a mock board rating lower than 34% of the total points. It is further predicted that if the general weighted average in all the general education subjects is fair, or the general weighted average in all the general education subjects is fairly good and has a kinesthetic learning style, then the reviewee will fail the LET.

No abstract

No abstract

No abstract

In training artificial neural networks, Backpropagation has been frequently used and known to provide powerful tools for classification. Due to its capability to model linear and non-linear systems, it is widely applied to various areas, offering solutions and help to human experts. However, BP still has shortcomings and a lot of studies had already been done to overcome it. But one of the important elements of BP, the stopping criterion, was given a little attention. The Fisher's Iris data set was used to this study as input for standard BP. Three experiments, using the different training set sizes, were conducted to measure the effectiveness of the proposed stopping criterion. The accuracy of the networks, trained in different data set sizes were also tested by using the corresponding testing sets. The experiments have shown that the proposed stopping criterion enabled the network to recognize its minimum acceptable error rate allowing it to learn to its maximum potential based on the presented patterns. The ubiquitous stopping criterion presented in this paper proved that the number of iterations to train the network should not be dictated by human since the accuracy of the network depends heavily on the number and quality of the training data.