RESEARCH ON FACTORS AFFECTING TOURIST INVOLVEMENT IN COFFEE TOURISM AFTER COVID-19 PANDEMIC IN THAILAND

In this paper, we analyze the factors affecting tourist involvement in coffee tourism after covid-19 pandemic in Thailand. The coffee tourism data is collected. Using Stop word removal, stemming, dimensionality reduction, Min-max normalization the data can be preprocessed. Feature can be extracted by using Linear Discriminant Analysis (LDA). Bat Algorithm is used for feature selection. Stochastic Neuro Fuzzy Decision Tree (SNF-DT) is used for Data analysis. Figure 2 indicates an overall methodology used. A. Dataset This study comprised 6,485,791 Thai and international visitors who visited Chiang Rai, Chiang Mai, Mae Hong Son, and Lampang provinces were included in this research. This study includes tourists from Thailand and abroad who visited coffee tourism destinations and participated in coffee-related tourism activities such as a tour of the cultivation and production area or a coffee tasting. The sample was selected using grab sampling. Based on ninety five percent confidences, the sample size was determined to be 10,000 participants with a margin of error of 0.05. The final result was 398, however it was rounded up to 400 to make data gathering. The proportion of people that responded is also shown for each province. Table 1 show that Sample Group in Each Province.

本研究针对新冠疫情（COVID-19 pandemic）后泰国游客参与咖啡旅游的影响因素展开分析。研究采集了咖啡旅游相关数据，通过停用词移除、词干提取、维度降维、最小-最大归一化完成数据预处理；采用线性判别分析（Linear Discriminant Analysis, LDA）提取特征，并以蝙蝠算法（Bat Algorithm）开展特征选择，最终使用随机神经模糊决策树（Stochastic Neuro Fuzzy Decision Tree, SNF-DT）进行数据分析。图2展示了本研究采用的整体研究流程。
A. 数据集
本研究纳入了共计6,485,791名赴泰国清莱府、清迈府、夜丰颂府及南邦府旅游的泰国及国际游客作为潜在研究对象。本次研究的分析对象为到访咖啡旅游目的地、参与咖啡种植生产区游览、咖啡品鉴等咖啡相关旅游活动的泰国及境外游客。样本采用便利抽样（grab sampling）方法选取。基于95%的置信水平与0.05的误差边际，本研究初始拟定样本量为10,000名参与者，最终实际回收有效样本398份，为便于后续数据归集工作，将其四舍五入至400份。各省份的受访者应答占比已予以展示，表1列出了各省份的样本分组情况。