ANALYSIS WINNOWING ALGORITHM FOR TEXT PLAGIARISM DETECTION USING THREE METHOD SIMILARITY

Cases of plagiarism in recent years has been an issues. Based on that issues, this research will create a system to detect similarity in a text. There is an aspect as reference of the research that is analyze the plagiarism algorithm. This research will analyze the accuracy one of plagiarism check algorithm, winnowing algorithm. Winnowing algorithm is a plagiarism detection algorithm based on document fingerprinting. To calculate percentage similarity of document fingerprinting in text, there are 3 methods to measure similarity that will be used in this research, which is jaccard similarity coefficient, sorensen dice similarity coefficient, and berg similarity coefficient.

A Novel Depth-Check Algorithm to Detect Macular Hole from Optical Coherence Tomography Images

Macular hole is a tear or opening forms in the macula. A macular hole forms a dark spot in the central vision and affects central vision, in this case the vision will be blurry, wavy or distorted. Macular hole commonly affects aged people. Optical coherence tomography enables accurate diagnosis of macular hole. Existing algorithms are also done related to finding layers, but macular hole identification in an accurate manner is still a missing entity. Hence we proposed a fully automated algorithm named “depth-check” for the accurate macular hole detection. The proposed method has six modules in process. First it starts with preprocessing the image, followed by nerve fiber layer (NFL) segmentation. The segmented image is then processed using depth-check algorithm. It will help to identify the macular hole from the optical coherence tomography images. For evaluation, we applied the algorithm on the optical coherence tomography images with the subjects- Central serous chorioretinopathy (CSCR) and Pigment epithelial detachment (PED). By experimentation, it is observed that the proposed algorithm provides 91% accuracy.