H. S. Anand, S. S. Vinod Chandra · 2013
Ever growing size of information and database has always demanded the scientific world for very efficient rule mining algorithm. This paper gives an extension to the Apriori algorithm, a classical rule mining algorithm. Apriori finds its application in areas of data mining, finding association between attributes and in prediction systems. Even though Apriori suits in various applications it possesses various disadvantages. To increase the efficiency of the present Apriori algorithm a method for incorporating a new correlation factor (threshold) is being introduced. First part of the paper provides a quick summary of basic Apriori algorithm and second half details the implementation of correlation threshold. Performance of the redesigned algorithm is evaluated and is compared with the traditional Apriori algorithm. The evaluation shows a peak improvement in the mining result. We reduce the time complexity of the newly designed algorithm into O (n). In an application level, qualitative content analysis of water was also conducted to affirm the results.
Cited by 17
Pei Yang, Wei Song, Xiaobing Zhao et al. · International Journal of Computational Science and Engineering · 2020
Image segmentation is widely used as a fundamental step for various image processing applications. This paper focuses on improving the famous image thresholding method named Otsu's algorithm. Based on the fact that threshold acquired by Otsu's algorithm tends to be closer to the class with larger intraclass variance when the foreground and background have large intraclass variance difference, an improved strategy is proposed to adjust the threshold bias. We analysed the relationship between pixel greyscale value and the change of cumulative pixel number, and selected the ratio of pixel grey level value to a certain cumulative pixel number as the adjusted threshold. Experiments using typical testing images were set up to verify the proposed method both quantitatively and qualitatively. Two widely used metrics named misclassification error (ME) and dice similarity coefficient (DSC) were adopted for quantitative evaluation, and both quantitative and qualitative results indicated that the proposed algorithm could better segment the testing images and get competitive misclassification error and DSC values compared with Otsu's method and its improved versions proposed by Hu and Gong (2009) and Xu et al. (2011), and the time consumption of our method can be significantly reduced.
Cited by 83
Lutfi Mohammed Omer Khanbary, Deo Prakash Vidyarthi · 2009
Genetic Algorithm (GA) is widely used for the number of optimization problems. For a large set of problems, GA provides the sub-optimal solution and thus the thrust to improve the GA for better result is on. In all the variants of the GA a selection operator guides the result towards the convergence. Selection, being a major component of the GA, attracts a good number of researchers with a new selection method. The current work introduces a variant of selection method in the GA that results in better improvement of the result. Genetic algorithm is modified by the new selection method that incorporates a problem specific knowledge in the new population generation. The GA with the new selection method is called as the Modified Genetic Algorithm (MGA). In the proposed selection method, not all the chromosomes of a generation are used for mating rather it is selected based on its goodness. Two threshold values, based on some prior knowledge, are used to classify the chromosomes. Efficacy of the proposed threshold method has been evaluated by conducting experiment and comparing the threshold method with some existing selection methods viz. sigma scaling, rank selection and tournament selection.
Cited by 7
Haoran Peng, Yuanfa Ji, Xiyan Sun et al. · 2023
To address the importance of code-deviation threshold selection in signal quality detection, a code-deviation-based threshold selection algorithm is proposed, firstly, a code-deviation threshold calculation model is given, a threshold selection method based on maximum likelihood estimation is derived, a small amount of abnormal data is eliminated by the box plot method, and then the mean and Then, the mean and standard deviation of the overall data are estimated by the maximum likelihood estimation method, and the thresholds that conform to the overall data distribution are calculated. The results show that the detection of code-carrier divergence can effectively identify the abnormal signals, prove the rationality of the threshold selection, solve the problem that the traditional threshold model is too conservative and less accurate, and improve the reliability of the threshold.
Cited by 0
Viet Ha−Thuc, Duc-Cuong Nguyen, Padmini Srinivasan · 2008
As database sizes increase, semantic data summarization techniques have been developed, so that data mining algorithms can be run on the summarized set for the sake of efficiency. Clustering algorithms such as K-Means have popularly been used as semantic summarization methods where cluster centers become the summarized set. The goal of semantic summarization is to provide a summarized view of the original dataset such that the summarization ratio is maximized while the error (i.e., information loss) is minimized. This paper presents a new clustering-based data summarization algorithm, in which the quality of the summarized set can be controlled. The algorithm partitions a dataset into a number of clusters until the distortion of each cluster is less than a given threshold, thus guaranteeing the summarized set has less than a fixed amount of information loss. Based on the threshold, the number of clusters is automatically determined. The proposed algorithm, unlike traditional K-Means, adjusts initial centers based on the information about the data space discovered so far, thus significantly alleviating the local optimum effect. Our experiments show that our algorithm generates higher quality clusters than K-Means does and it also guarantees an error bound, an essential criterion for data summarization.
Cited by 13
L.V. Baskakova · USSR Computational Mathematics and Mathematical Physics · 1980
Cited by 1
Elise Dusseldorp, Claudio Conversano · 2013
Regression trunk model estimation proposed by Dusseldorp and Meulman (2004) <<a href="https://doi.org/10.1007%2Fbf02295641" target="_top">doi:10.1007/bf02295641</a>> and Dusseldorp, Conversano, Van Os (2010) <<a href="https://doi.org/10.1198%2Fjcgs.2010.06089" target="_top">doi:10.1198/jcgs.2010.06089</a>>, integrating a regression tree and a multiple regression model.
Cited by 1Open access
T. Kawai, Ryoko Nakano · 2005
The EM algorithm is an efficient algorithm to obtain the ML estimate for incomplete data, but has the local optimality problem. The deterministic annealing EM (DAEM) algorithm was once proposed to solve the problem, but the global optimality is not guaranteed because of a single-token search. Then, the multi-thread DAEM (m-DAEM) algorithm was proposed by incorporating a multiple-token search with solution quality improvement with a heavy computing cost. Later, another variant of m-DAEM (/spl epsiv/-DAEM) was proposed by introducing threshold-based dynamic annealing with more quality improvement for an adequate threshold /spl epsiv/; however, finding such /spl epsiv/ is not easy. This paper proposes a new variant of EM, called /spl epsiv/-EM, by incorporating a multiple-token search together with threshold-based bifurcation. Our experiments using Gaussian mixture estimation problems showed that the /spl epsiv/-EM finds excellent solutions with relatively small computing cost, and the threshold /spl epsiv/ plays a key role in reducing computing cost.
Cited by 2
Wenqing Chen Wentan Jiao · CONVERTER · 2021
The Improved Bat Algorithm (IBA) is proposed for the image segmentation based on the maximum interclass variance method. Firstly, the principle of image segmentation based on the maximum interclass variance method is explained, and secondly, the bat algorithm is improved by using chaotic logistic mapping to initialize the population to improve the diversity of solutions, using adaptive parameter optimization to avoid falling into local optimum, using Monkey algorithm for individual selection, and finally, the image segmentation function in image segmentation is used as the individual fitness function of the bat algorithm for solving. The simulation experiments show that compared with the bat algorithm and the monkey group algorithm, this algorithm has better segmentation effect under different threshold values.
Cited by 1Open access
G. Chandra Sekhar · Ophthalmology · 2000
Cited by 82
W.S. Li, Yu Cao · WIT transactions on information and communication technologies · 2014
Welcome to the WIT Press eLibrary - the home of the Transactions of the Wessex Institute collection, providing on-line access to papers presented at the Institute's prestigious international conferences and from its State-of-the-Art in Science & Engineering publications.
Cited by 0
S. Götz, Philip Whiting · 2011
Cited by 0
J.P. Reilly, Xin Huang, K.M. Wong · Canadian Journal of Electrical and Computer Engineering · 1989
Two methods are presented for extending the threshold characteristics of the modified forward-backward linear prediction (MFBLP) algorithm due to Tufts and Kumaresan (1982). This algorithm estimates angles of arrival of plane waves onto linear arrays of sensors. The first technique proposed, referred to as modified-modified FBLP (M <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> FBLP), offers 4- to 6-dB threshold extensions over MFBLP for a large number of snapshots, for a particular set of simulation parameters. The second technique, called optimized FBLP (opt-FBLP), offers similar threshold extension when the numer of snapshots is low. The opt-FBLP method is most successful when the number of incident signal components <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> is 2. The method is therefore viewed as being particularly applicable to the low-angle tracking problem in radar, since it has been shown that there is strong justification to fix the value of <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> at 2 in this situation. The increase in computational complexity required for M <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -FBLP over MFBLP is virtually negligible. For opt-FBLP, the increase in computational complexity is minimal for <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> =2.
Cited by 0
2008
Cited by 0
N. Sharma, P.Thumm, D.Saupe · 2014
Cited by 0Open access
A.N.M. Rezaul Karim, Masidur Alam · Journal of Convergence Information Technology · 2009
Cited by 4
Hao Qiu, Shuaishuai Peng · 2022
The adaptive threshold based SIFT image registration algorithm (AT-SIFT) is proposed to improve the traditional SIFT algorithm which has the problems of large calculation amount, single matching constraint, and low registration accuracy. Feature points are detected and described by combining FAST operators with and SIFT descriptors. The FLANN algorithm based on adaptive threshold and the improved RANSAC algorithm are used to remove the mis-matching points and solve the transformational matrix. The experimental results show that the algorithm not only improves the matching speed and accuracy but also has relatively strong robustness.
Cited by 3
Shao‐Li Chen, Min Yang · 2017
Fast iterative shrinkage threshold algorithm (FISTA) is an efficient first-order optimization algorithm for Linear inverse problems. However, the algorithm employed a fixed iterative step size which limits the speed of calculation, This paper proposes an adaptive fast iterative shrinkage threshold algorithm (FISTA) by using a Barzilai-Borwein (BB) operator. The proposed Algorithm uses the previous iteration information to update the step size which can speed up the rate of the iteration. The numerical experimental results of Compressed Sensing and Image Denoising demonstrate that the proposed algorithm has a faster convergence rate and improves the efficiency of the calculation.
Cited by 5
Я.А. Буянкина, П.В. Бабаян · 2025
В работе рассматривается актуальная проблема сегментации изображений, полученных в условиях наблюдения, характеризующихся мультимодальностью гистограммы объекта. Для решения этой проблемы используется генетический алгоритм. В рамках исследования представлена реализация алгоритма, а также проведены эксперименты с различными изображениями, что позволяет оценить его эффективность. Результаты экспериментов включают в себя расчёт характеристик, которые демонстрируют улучшение качества сегментации по сравнению с традиционными методами, такими как метод Отцу и энтропийные методы Капура и Тсаллиса. The paper considers the current problem of segmentation of images obtained under observation conditions characterized by multimodality of the object histogram. To solve this problem, a genetic algorithm is used. The study presents the implementation of the algorithm, as well as experiments with various images, which allows us to evaluate its effectiveness. The results of the experiments include the calculation of characteristics that demonstrate an improvement in the quality of segmentation compared to traditional methods such as the Otsu method and the entropy methods of Kapur and Tsallis.
Cited by 0Open access
Wei Han, Zuyuan Yang, Jun Lu et al. · Physiological Measurement · 2018
Our study develops a methodology to determine the category of a predicted heart sound instance from its segments' prediction results, thus assisting in the data augmentation exercise which is necessary to provide sufficient data for deep classification networks. Our method significantly improves the classification performance.
Cited by 33