Atchariyachanvanich, Kanokwan

Preferred name

Main Affiliation

School of Information Technology

kanokwan.at@kmitl.ac.th

Now showing 1 - 10 of 33

Some of the metrics are blocked by your
consent settings
RSQLG: The Reverse SQL Question Generation Algorithm
(2019-05-14)
Julavanich, Thanakrit
;
Nalintippayawong, Srinual
;
Atchariyachanvanich, Kanokwan
Nowadays, the people who learn and teach SQL commands require to get a hand-on practice with real environment to make the learning effective. The creating SQL exercise is a time-consuming task for instructors. As a result, students might not get enough questions to meet their demands. The reverse SQL question generation algorithm (RSQLG) is developed to solve this problem. The RSQLG has ability to automatically generate SQL exercise for students. The RSQLG can reverse the manual question creation process which starts from creating question to start creating query answer first instead. The RSQLG considers the existing data and database structure by using various constraints. The instructors also can specify the language, format and explanation of the questions. The RSQLG supports DML commands-SELECT, INSERT, UPDATE and DELETE-and support data retrieval from multiple tables which perform by JOIN and subquery operations. The algorithm has ability to generate bulk questions with less effort. The instructor is not required to writing requirements and validating the queries. The RSQLG can be implemented in e-learning to enhance sustainable practices and improve learning outcome for students.
Some of the metrics are blocked by your
consent settings
Improvement of a Machine Learning Model Using a Sentiment Analysis Algorithm to Detect Fake News: A Case Study of Health and Medical Articles on Thai Language Websites
(2024-01-01)
Atchariyachanvanich, Kanokwan
;
Saengkhunthod, Chotipong
;
Kerdnoonwong, Parischaya
;
Chanlekha, Hutchatai
;
Cooharojananone, Nagul
These days, the problem of fake news has grown to be a major social and personal concern. With the amount of information generated through social media, it is very crucial to be able to detect and properly take care of that fake information. Previous studies proposed a machine learning model to detect fake news in online Thai health and medical articles. Still, the problem of detecting fake news with similar content but different objectives exists, and the accuracy of the model needs improvement. Therefore, this study aims to solve these problems by adding 33 new features, including textual features, sentiment-based features, and lexicon features, i.e., herbs, fruits, and vegetables, to identify the objective of an article. We trained and tested the model’s prediction accuracy on a new dataset containing 582 reliable and 435 unreliable (fake news) articles from eight Thai websites. Our improved classification model using XGBoost with Lasso, the best feature selection method, achieved an accuracy of 97.76% without over-fitting, reflecting a 7.16% improvement over our earlier model.
Some of the metrics are blocked by your
consent settings
The study of the local community products (OTOP) website characteristics toward buyer decision using eye tracking
(2018-05-08)
Cooharojananone, Nagul
;
Akasarakul, Krittika
;
Wongkhamdi, Thipsuda
;
Pruetthiwongwanich, Phonkornkrit
;
Atchariyachanvanich, Kanokwan
Thai government has launched a policy to encourage local community people to sell their products (OTOP) through e-commerce. One Tambon One Product or OTOP was derived from the concept of One Village One Product (OVOP) in Japan. Currently, people sell their products through the e-commerce web portal such as Facebook rather than having their own OTOP official websites. Therefore, this study aims to study and develop the prototype of the official website. The results show that social influence, perceived ease of use, reliability of website and reliability of product have a positive relationship to the intention to buy a product. Researchers further investigated using eye tracking device to analyze the participants' purchasing behavior. The results show that participants spent time looking at images on the center of the homepage, the product menu bar at the bottom of the center of the homepage, and the components from the product page, respectively. Moreover, participants did not spend time looking or even clicking the social network icon such as Facebook. The results of this study and implications are further discussed.
Some of the metrics are blocked by your
consent settings
An exploratory study on users' intention to purchase items from mobile applications in Thailand
(2015-10-05)
Atchariyachanvanich, Kanokwan
;
Nararatwong, Rungsiman
;
Cooharojananone, Nagul
In this paper, we study the relationship between continuation of use (COU) factor and the users' intention to purchase items in smartphone applications. We also proposed the four factors: trust, diversity of in- app items, social norm and resource usage, specifically affect the COU of the smartphone applications. Structural equation model (SEM) analysis of a questionnaire based survey collected from 226 respondents in Thailand revealed that the continual use of an application is the key factor affecting the intention of users to make in-app purchases. This factor is affected by the diversity of in-app items, trust and social norms. However, the cost of using the application does not significantly affect the continuation of use. This study contributes to m-commerce as we proposed the analysis model, which specifically adapted to smartphone applications targeting consumers in Thailand. The result helps developers to focus on crucial factors which will enhance the possibility of users' continuous use and decision to purchase in-app items.
Some of the metrics are blocked by your
consent settings
Comparative Study of Prediction Models for Final GPA Score: A Case Study of Rajabhat Rajanagarindra University
(2018-09-14)
Putpuek, Narongsak
;
Rojanaprasert, Natcha
;
Atchariyachanvanich, Kanokwan
;
Thamrongthanyawong, Thananya
Recently, the analysis of educational data has become important to all universities. Rajabhat Rajanagarindra University, Thailand, wanted to study and analyze the students' performance based on their personal background. Thus, this research aimed to compare prediction models for the level of the final grade point average (GPA) score of graduated students using the data from the Faculty of Education during the 2010 to 2012 academic years. Two decision tree (C4.5 and ID3) algorithms, plus Naïve Bayes and K-nearest neighbor data mining techniques were adopted to analyze the data according to the CRISP-DM process. Factors that were proposed to influence the graduation GPA include the student's gender, scholarship awarded, previous educational background, admission type, talent and province of high school. The analysis revealed that the Naïve Bayes algorithm gave the best overall accuracy of 43.18%. This could help predict the graduation GPA score of students in the future and support teachers to make educational advice for their students and to develop the student quality in the future.
Some of the metrics are blocked by your
consent settings
Reverse SQL question generation algorithm in the dblearn adaptive e-learning system
(2019-01-01)
Atchariyachanvanich, Kanokwan
;
Nalintippayawong, Srinual
;
Julavanich, Thanakrit
Using a traditional e-learning system, when teaching structured query language (SQL) queries in classical classrooms help instructors, to improve the students' SQL skills and learning effectiveness. However several problems in using e-learning as a teaching and learning assistant remain-such as difficulties in differences in learning ability and knowledge level. We solved these problems by applying an adaptation module to our e-learning system. However, we still found it required considerable effort to create enough exercises to make the adaptation effective enough. So, we developed a novel automatic question generating algorithm, named Reverse SQL Question Generation Algorithm (RSQLG), to automatically generate exercises (including both answer and question) from a source database. RSQLG reverses the traditional manual process used previously by instructors. Instead of creating questions and answers for them, RSQLG creates the answers first. The generated exercises are presented to students by applying question adaptation methodology based on student knowledge level in each supported learning objective. We evaluated the learning effectiveness of our approach by using outcome-based learning. After post-Test to pre-Test scores were compared, we found students using our system improved their scores by 26%. Consequently, the adaptive e-learning framework using RSQLG could be applied in any adaptive or traditional e-learning for a database course to benefit the instructors leading to less effort in exercise management and to improve the learning outcome from the students allowing as much practice as they need.
Some of the metrics are blocked by your
consent settings
Factors affecting users' attitude toward Facebook application in Thailand
(2013-04-22)
Nararatwong, Rungsiman
;
Pongsupankij, Nattarat
;
Atchariyachanvanich, Kanokwan
;
Cooharojananone, Nagul
Since Facebook was launched, privacy related issues had been discussed intensively. Facebook's third-party applications became a huge challenge in maintaining user's privacy since user's information can be shared to the developers outside Facebook. This paper studies factors that affect user's attitude on using Facebook application. Based on our hypothesis, three factors are: brand loyalty, user's self benefit and user's social benefit. We conducted an online survey which collected opinions from 246 respondents in Thailand. Respondents were asked to answer the online questionnaire twice, once before and again after reading the story about risk of information disclosure. We performed a regression analysis to test our hypotheses and the result showed that before reading the story, only self benefit affects user's attitude. However, after reading the story, brand loyalty becomes the only factor that user takes into consideration. © 2013 GIRI.
Some of the metrics are blocked by your
consent settings
IT management status in public higher education institutions in Thailand
(2016-08-23)
Nalintippayawong, Srinual
;
Atchariyachanvanich, Kanokwan
Thai public higher education institutions have adopted information technology (IT) to some extent for many years. While IT adoption is an important strategy to make any organization competitive against others, there has been very little research on IT management issues and the depth of use of IT across Thai public higher education institutions. Therefore, this study aimed to explore the status of information system adoption in public higher education institutions in Thailand, and to analyze the strengths, weaknesses, opportunities and threats in the adoption of IT. A questionnaire survey was sent to chief information officers (CIOs) or equivalent position of 81 target Thai public higher education institutions. The 63 valid responses (77.8%) received were analyzed by qualitative content analysis and descriptive statistics. This research outcome will reflect the overall IT adoption and IT management of Thai public higher education institutions.
Some of the metrics are blocked by your
consent settings
Interactive SQL Learning Tool with Automated Grading using MySQL Sandbox
(2019-05-14)
Permpool, Tanasab
;
Nalintippayawong, Srinual
;
Atchariyachanvanich, Kanokwan
This research presents an original tool for automatic grading of SQL statements using MySQL Sandbox[l]. This tool allows students to create and execute the queries in real time. There are many ways to write accurately SQL statement for one question. This is a pain point and burden to the instructors. The existing automated SQL grading algorithms have limitations on supportable statements and understandable error notification. The existing automated SQL grading tool did not support grading on data manipulation language statements i.e. INSERT, UPDATE and DELETE and do not support data definition language commands also; i.e. CREATE, ALTER and DROP. This paper presented an original algorithm which can support these commands. The proposed algorithm can analyze and display the differences between student's answer and correct answer. This algorithm can reduce the instructors' burdens by automatic scoring students' answers and the results directly notify back to both instructors and students. The students can learn and improve themselves faster and effectively.
Some of the metrics are blocked by your
consent settings
Development of a MySQL Sandbox for processing SQL statements: Case of DML and DDL statements
(2017-09-05)
Atchariyachanvanich, Kanokwan
;
Nalintippayawong, Srinual
;
Permpool, Tanasab
This research developed the MySQL Sandbox, a secured environment for processing SQL queries. It was implemented as a RESTful web service having three services-sandbox database creation, SQL statement processing and sandbox database resetting. It supports the simultaneous processing of multiple SQL statements from multiple users in multiple databases. It uses question identification (ID) and student ID to create separate databases for each student using the MySQL feature to manage the user's privileges of their own database. Every service returns a result in the JSON format, which is easy to understand. This MySQL Sandbox is the first tool to support judging DDL statements and complex DML statements. Existing SQL grading systems have limitations on the number of supported SQL statements because they are concerned about risks from some sensitive SQL statement, such as DDL and DML statements, other than the SELECT statement. This sandbox will help eliminate the security concerns that obstruct the development and improvement of SQL grading systems, while providing a greater freedom of learning query to students, which will help them improve their own skills in three dimensions i.e., database query, database administration and database programming.

Atchariyachanvanich, Kanokwan

Filters

Settings

Sort By

Results per page