KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 27

Some of the metrics are blocked by your
consent settings
A Study of Using GPT-3 to Generate a Thai Sentiment Analysis of COVID-19 Tweets Dataset
(2023-01-01)
Isaranontakul, Patthamanan
;
Kreesuradej, Worapoj
This study evaluated the effectiveness of using synthetic text datasets generated by GPT-3 for sentiment analysis with deep learning models, namely Bi-GRU and Bi-LSTM. The study compares the performance of these models on both synthetic text and Label Tweet datasets using GPT-3 and reveals that deep learning model performance is dependent on the dataset's nature. The results indicate that using synthetic text generated by GPT-3 significantly enhances the accuracy of both models, with Bi-LSTM achieving an accuracy of 0.84 and Bi-GRU achieving an accuracy of 0.85. The study underscores the importance of meticulous dataset selection and preparation for developing precise and effective deep learning models for various sequential data types. The findings demonstrate that synthetic text datasets generated by GPT-3 can serve as a valuable resource for developing deep learning models as they are labeled and save researchers time and effort in manual labeling of large datasets.
Some of the metrics are blocked by your
consent settings
Incremental Association Rule Mining with a Fast Incremental Updating Frequent Pattern Growth Algorithm
(2021-01-01)
Thurachon, Wannasiri
;
Kreesuradej, Worapoj
One of the most challenging tasks in association rule mining is that when a new incremental database is added to an original database, some existing frequent itemsets may become infrequent itemsets and vice versa. As a result, some previous association rules may become invalid and some new association rules may emerge. We designed a new, more efficient approach for incremental associationrule mining using a Fast Incremental Updating Frequent Pattern growth algorithm (FIUFP-Growth), a new Incremental Conditional Pattern tree (ICP-tree), and a compact sub-tree suitable for incrementalmining of frequent itemsets. This algorithm retrieves previous frequent itemsets that have already been mined from the original database and their support counts then use them to efficiently mine frequent itemsets from the updated database and ICP-tree, reducing the number of rescans of the original database. Our algorithm reduced usages of resource and time for unnecessary sub-tree construction compared to individual FP- Growth, FUFP-tree maintenance, Pre-FUFP, and FCFPIM algorithms. From the results, at 3% minimum support threshold, the average execution time for pattern growth mining of our algorithm performs 46% faster than FP- Growth, FUFP-tree, Pre-FUFP, and FCFPIM. This approach to incremental association rule mining and our experimental findings may directly benefit designers and developers of computer business intelligence methods.
Some of the metrics are blocked by your
consent settings
The framework of government cloud computing adoption with TAM in Thailand
(2019-08-23)
Buavirat, Warune
;
Kreesuradej, Worapoj
;
Chaveesuk, Singha
Cloud technology/ Cloud Computing had gained significance and popularity in the recent years with the increase in the internet reliability, even increasing access speed and the need for large storage requirements by various users. The cloud technology has been widely used in the various business industries for many years such as data storage, analytics, data management, connecting network, distributed working etc. Cloud computing offer its series of benefits to many business industries, organizations including at government level. The implication of Cloud technology in the government will help to get information in real-time, improve work process with higher efficiency and effectiveness. E-government also known as G-Cloud is the kind of cloud system, which provides user interface to the Government systems for its citizens. The influx of innovation had also brought up many disruptions to many businesses in the industry. Organizations have been involved in the research to overcome such challenges and aims to extend adoption and technology fit with better clarity while overcoming the disruptions that have been faced by its citizens from the government departments and why it’s necessary for adopting this technology innovation by their particular organization. This study based on technology acceptance model 3 and task-technology fit model.
Some of the metrics are blocked by your
consent settings
A Technique for Estimating Updated Frequent Itemsets in ESC-Growth Algorithm
(2019-07-01)
Kreesuradej, Worapoj
;
Thurachon, Wannasiri
In discovering association rules from a dynamic database, iteration through the frequent itemsets requires significant resources and computational time for construction of sub-trees, sub-tree traversal and generation of conditional pattern bases, and it is quite possible that no updated frequent itemsets may have been found at all, resulting in a waste of resources and computational time. We describe a technique for estimating the support count for the itemsets for the next iteration of discovery of the frequent itemsets by our ESC-Growth Algorithm. This technique reduces the need to construct a new sub-tree and next discovery step. If no frequent itemsets in the updated database have been found in the next iteration, ESC-Growth will not construct a new sub-tree and will stop discovering new frequent itemsets in that iteration, reducing the waste of resources and computational time. We measured execution time and sub-tree counts for FP-Growth, FUFP-tree, FPISC-Growth and ESCGrowth on the same synthetic dataset; we found that, at 5% minimum support threshold, ESC-Growth used only 40.3, 96.5, and 99.6% of the execution time required by FP-Growth, FUFPtree and FPISC-Growth, respectively.
Some of the metrics are blocked by your
consent settings
Webcam based eye gaze prediction system with automatic calibration for web browser
(2019-07-01)
Karngumpol, Niphat
;
Kreesuradej, Worapoj
Now a day, researches about Eye Gaze Prediction Systems with General Webcam was interested by many researchers and has been continuously developed because it is cheap and can be applied in many ways. Currently, a Browser based open source library for eye gaze prediction has been developed, which is easy to use and can be developed in a variety fields. However, this method is still limited because of low precision and inconveniences because users must perform calibration every time before use it. Therefore, we propose ways to develop and solve the above problems. This research has two objectives. The first is to improve the accuracy of an existing Eye Gaze Prediction System. The solution is using the Simple Moving Average to reduce the volatile of results and increase accuracy. From a results, this method gives a test scores higher than the existing method with statistically significant at 0.01, calculated as 14.96 percent increase from average score of the old method. The second objective is to propose a solution for making the system can recalibrate itself to improve accuracy in a long time without having to perform calibration by the users. We record the gaze data from system in the first 1 minute and then take the recorded data to calculate the boundary of the screen. And then, compare the calculated boundary with the real screen boundary to find error factors. This calculated error factor has been applied back to the next result to increase accuracy. From the test results, we found that over time the average test scores gradually increase. And at the last minute, the average score from this method is higher than the average score from traditional method up to 13.66 percent.
Some of the metrics are blocked by your
consent settings
Discovery of incremental association rules based on a new FP-growth algorithm
(2019-02-01)
Kreesuradej, Worapoj
;
Thurachon, Wannasiri
In this paper, we propose a new FP-Growth algorithm for incremental association rule discovery. We also design a new FPISC-tree based on the FUFP-tree structure. The new FPISC-tree is more suitable for the task of incremental association rule discovery than FUFP-tree structure. The basic ideas of the proposed algorithm are to retrieve the frequent itemsets from the original database and to use their support count in the update of the new support count of the incremental database so that the original paths do not need to be reprocessed as well as to strategically use them to discover frequent itemsets from the FPISC-tree. Experimental results show that the proposed algorithm was able to reduce the number of constructed subtrees and the execution time was significantly less than those of the FP-Growth and FUFP-tree.
Some of the metrics are blocked by your
consent settings
The acceptance model of M-payment using smartphone devices in Thailand: A conceptual framework
(2018-01-01)
Nakwari, Parisgawin
;
Kreesuradej, Worapoj
;
Chaveesuk, Singha
In this paper, we have proposed the M-payment method using the smartphone applications, from which all electronic transactions such as purchasing, sales and the provision of other financial services can be applied. This is the investigation of the UTAUT2 extended theory that can be more appropriate for the consumers. The UTAUT2 extended theory is integrated with the service quality, which can be used to examine the acceptance potential of the M-Payment user by smartphone devices. The outcomes of this study will benefit the financial industry in their effort to initiate an extensive online-payment process in all users, especially, in Thailand.
Some of the metrics are blocked by your
consent settings
Frequent itemsets mining using random walks for record insertion and deletion
(2017-02-23)
Thusaranon, Panita
;
Kreesuradej, Worapoj
In Association rules mining, the task of finding frequent itemsets in dynamic database is very important because the updates may not only invalidate some existing rules but also make other rules relevant. In this paper, we propose a new algorithm to maintain frequent itemsets of a dynamic database in the case of record insertion as well as deletion simultaneously. Basically, the proposed algorithm maintains not only the support counts of frequent itemsets but also the support counts of prospective frequent itemsets, i.e., infrequent itemsets that promise to be frequent in the future, in an original database. Prospective frequent itemsets, which are obtained by using the principle of Random Walks, can help to reduce a number of times to rescan the original database.
Some of the metrics are blocked by your
consent settings
An enhanced incremental association rule discovery with a lower minimum support
(2016-12-01)
Ariya, Araya
;
Kreesuradej, Worapoj
In the real world of data, a new set of data has been being inserted into the existing database. Thus, the rule maintenance of association rule discovery in large databases is an important problem. Every time the new data set is appended to an original database, the old rule may probably be valid or invalid. This paper proposed the approach to calculate the lower minimum support for collecting the expected frequent itemsets. The concept idea is applying the normal approximation to the binomial theory. This proposed idea can reduce a process of calculating probability value for all itemsets that are unnecessary. In addition, the confidence interval is also applied to ensure that the collection of expected frequent itemsets is properly kept.
Some of the metrics are blocked by your
consent settings
Maintenance of multi-level association rules discovery in dynamic database under a change of support threshold
(2016-01-13)
Pumjun, Nophadon
;
Kreesuradej, Worapoj
An association rule mining is often performed with a dynamic database and hierarchical items. The big problem of data mining process is a maintenance association rules while the database always changing. The purpose of this study is to extend the MLUp algorithm which can maintain a multilevel association rules discovery at the same minimum support threshold. In general, several mining tasks are required to deal with different support thresholds. MLUpCS can deal with a maintaining of mining multilevel association rules in dynamic databases under the different support threshold without re-mine a whole database. The result of MLUpCS algorithm experiment has shown how better performance than ML-T2 algorithm. The experimental results show the superior performance of MLUpCS when compared with ML-T2.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results