Intakosum, Sarun

Preferred name

Intakosum, Sarun

Main Affiliation

School of Science

sarun.in@kmitl.ac.th

Now showing 1 - 5 of 5

Some of the metrics are blocked by your
consent settings
An efficient parallel construction of optimal independent spanning trees on hypercubes
(2012-12-01)
Werapun, Jeeraporn
;
Intakosum, Sarun
;
Boonjing, Veera
Reliable data broadcasting on parallel computers can be achieved by applying more than one independent spanning tree (IST). Using k-IST-based broadcasting from root r on an interconnection network (N=2 k) provides k-degree fault tolerance in broadcasting, while construction of optimal height k-ISTs needs more time than that of one IST. In the past, most research focused on constructing k ISTs on the hypercube Qk, an efficient communication network. One sequential approach utilized the recursive feature of Qk to construct k ISTs working on a specific root (r)=0 in O(kN) time. Another parallel approach was introduced for generating k ISTs with optimal height on Qk, based on HDLS (Hamming Distance Latin Square), single pointer jumping, which is applied for a source (r)=0 in O( k2) time for successful broadcasting in O(k). For broadcasting from r≠0, those existing approaches require a special routine to reassign new nodes' IDs for logical r=0. This paper proposes a flexible and efficient parallel construction of k ISTs with optimal height on Qk, a generalized approach, for an arbitrary root (r=0,1,2,..., or 2 k-1) in O(k) time. Our focus is to introduce the more efficient time (O(k)) of preprocessing, based on double pointer jumping over O( k2) of the HDLS approach. We also prove that our generalized parallel k-IST construction (arbitrary r) with optimal height on Qk is correctly set in efficient O(k) time. Finally, experiments were performed by simulation to investigate the fault-tolerance effect in reliable broadcasting. Experimental results showed that our efficient ISTs yielded 10%-20% fault tolerance for successful broadcasting (on N=16-1024 PEs). © 2012 Elsevier Inc. All rights reserved.
Some of the metrics are blocked by your
consent settings
High candidates generation: A new efficient method for mining share-frequent patterns
(2017-11-01)
Nawapornanan, Chayanan
;
Intakosum, Sarun
;
Boonjing, Veera
The share frequent patterns mining is more practical than the traditional frequent patternset mining because it can reflect useful knowledge such as total costs and profits of patterns. Mining share-frequent patterns becomes one of the most important research issue in the data mining. However, previous algorithms extract a large number of candidate and spend a lot of time to generate and test a large number of useless candidate in the mining process. This paper proposes a new efficient method for discovering share-frequent patterns. The new method reduces a number of candidates by generating candidates from only high transaction-measure-value patterns. The downward closure property of transaction-measure-value patterns assures correctness of the proposed method. Experimental results on dense and sparse datasets show that the proposed method is very efficient in terms of execution time. Also, it decreases the number of generated useless candidates in the mining process by at least 70%.
Some of the metrics are blocked by your
consent settings
Closed multidimensional sequential pattern mining
(2006-11-14)
Songram, Panida
;
Boonjing, Veera
;
Intakosum, Sarun
We propose a new method, called closed multidimensional sequential pattern mining, for mining multidimensional sequential patterns. The new method is an integration of closed sequential pattern mining and closed itemset pattern mining. Based on this method, we show that (1) the number of complete closed multidimensional sequential patterns is not larger than the number of complete multidimensional sequential patterns (2) the set of complete closed multidimensional sequential patterns covers the complete resulting set of multidimensional sequential patterns. In addition, mining using closed itemset pattern mining on multidimensional information would mine only multidimensional information associated with mined closed sequential patterns, and mining using closed sequential pattern mining on sequences would mine only sequences associated with mined closed itemset patterns. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
Artificial Neural Network and Genetic Algorithm Hybrid Intelligence for Predicting Thai Stock Price Index Trend
(2016-01-01)
Inthachot, Montri
;
Boonjing, Veera
;
Intakosum, Sarun
This study investigated the use of Artificial Neural Network (ANN) and Genetic Algorithm (GA) for prediction of Thailand's SET50 index trend. ANN is a widely accepted machine learning method that uses past data to predict future trend, while GA is an algorithm that can find better subsets of input variables for importing into ANN, hence enabling more accurate prediction by its efficient feature selection. The imported data were chosen technical indicators highly regarded by stock analysts, each represented by 4 input variables that were based on past time spans of 4 different lengths: 3-, 5-, 10-, and 15-day spans before the day of prediction. This import undertaking generated a big set of diverse input variables with an exponentially higher number of possible subsets that GA culled down to a manageable number of more effective ones. SET50 index data of the past 6 years, from 2009 to 2014, were used to evaluate this hybrid intelligence prediction accuracy, and the hybrid's prediction results were found to be more accurate than those made by a method using only one input variable for one fixed length of past time span.
Some of the metrics are blocked by your
consent settings
Predicting SET50 index trend using artificial neural network and support vector machine
(2015-01-01)
Inthachot, Montri
;
Boonjing, Veera
;
Intakosum, Sarun
Finding ways to making better prediction of stock market trend has attracted a lot of attention from researchers because an accurate prediction can substantially reduce investment risk and increase profit gain for investors. This study investigated the use of two machine learning methods, Artificial Neural Network (ANN) and Support Vector Machine (SVM), for predicting the trend of Thailand’s emerging stock market, SET50 index. Raw SET50 index records from 2009 to 2013 were converted into 10 widely-accepted technical indicators that were then used as input for model construction and testing. Our test results showed that the accuracy of the ANN model outperforms that of the SVM model.

Intakosum, Sarun

Filters

Settings

Sort By

Results per page