澎湖擁有豐富且獨特的自然和文化資源，這些資源是該地區觀光發展的潛力所在，也是吸引旅客前來的主要原因。根據民國110年4月核定的「澎湖縣國土計畫」，澎湖縣政府提出了以「海洋生態、創意觀光、序續菊島」作為空間發展的願景，目標是建立一個結合低碳生態保育和創意觀光的有序與永續菊島。從合法民宿家數的增長中，我們可以見證地方觀光產業的發展。澎湖的觀光成果，不僅來自其天然的條件，還包括政府的政策導向和民間業者的持續努力。然而，觀光產業是一個極易受到外部環境影響的產業，如2019年底爆發的Covid-19疫情對全球觀光業造成了巨大衝擊。
在行動通訊和社群媒體廣泛普及的當下，社群媒體為旅客提供了一個分享旅遊訊息和體驗的平臺，這些由使用者生成的內容，能更真實地反映旅客的感受和想法。在自願性地理資訊和適地性社群網絡分析的應用下，除了能夠彌補官方資料之不足，許多情況下更能夠反映旅遊景點真實且細微情形，從而得知城市中的活動以及遊客的行為，進而協助政府優化資料收集和政策制定方向。因此，本研究根據「澎湖觀光發展整體規劃」，利用社群媒體Instagram打卡貼文數據，選取10個遊憩主題次系統內45處具有代表性的旅遊景點作為依據，研究旨在利用Instagram對澎湖各著名旅遊景點進行遊客人數推估，以彌補官方統計資料之不足，並探討疫情對澎湖觀光旅遊人潮的影響與各景點的吸睛熱度，以及提供政府觀光旅遊及外在環境因素衝擊下之應對參考。
研究結果顯示，以Instagram主題標籤貼文(遊客)數進行觀光旅遊人次推估，適用於沒有官方遊客統計的戶外景點採用，其迴歸模型的配適度佳;然而，對於室內景點，此方法則不適用，因其迴歸模型的配適度極差。再從Instagram打卡貼文(遊客)數來看，「摩西分海(奎壁山)」、「跨海大橋」、「雙心石滬」、「篤行十村」、「隘門沙灘」、「山水沙灘」、「鯨魚洞」、「風櫃洞」及「大菓葉柱狀玄武岩」這九個景點不論是平常，亦或是疫情間依然保持高吸睛程度。另外景點「後寮天堂路」在109年3月至8月期間，Instagram打卡貼文(遊客)數擠進十大吸睛景點內，且其景點特性與吸睛榜首「摩西分海(奎壁山)」相同，必需掌握潮汐時間，方能體現該景點的奧秘，是個極具發展淺力旅遊景點。

Penghu possesses rich and unique natural and cultural resources, which are the potential for the region's tourism development and the main reason attracting visitors. According to the "Penghu County National Land Plan" approved in April 2021, the Penghu County government proposed a spatial development vision of "marine ecology, creative tourism, and the sustainable development of Penghu," aiming to establish an orderly and sustainable Penghu Island combining low-carbon ecological conservation and creative tourism. The growth in the number of legal B&Bs reflects the development of the local tourism industry. The achievements in Penghu's tourism not only come from its natural conditions but also include government policy orientations and the continuous efforts of private operators. However, the tourism industry is highly susceptible to external environmental influences, such as the significant impact of the Covid-19 pandemic that broke out at the end of 2019 on the global tourism industry.
With the widespread popularity of mobile communications and social media, social media provides a platform for travelers to share travel information and experiences. User-generated content can more authentically reflect travelers' feelings and thoughts. Through the application of Volunteered Geographic Information (VGI) and Location-Based Social Network (LBSN) analysis, it is possible to compensate for the shortcomings of official data and, in many cases, reflect the real and nuanced situations of tourist attractions, thereby understanding activities in the city and tourists' behavior, which in turn helps the government optimize data collection and policy-making directions. Therefore, this study, based on the "Overall Planning for Penghu Tourism Development," uses Instagram check-in post data to select 45 representative tourist attractions within 10 recreational thematic subsystems. The study aims to use Instagram to estimate the number of visitors to various famous tourist spots in Penghu, to compensate for the lack of official statistical data, and to explore the impact of the pandemic on tourist flows in Penghu and the attractiveness of various attractions. It also aims to provide references for government tourism and responses to external environmental factors.
The results show that estimating the number of tourists based on Instagram hashtag posts is suitable for outdoor attractions without official tourist statistics, with a good fit for the regression model; however, this method is not applicable to indoor attractions due to the poor fit of their regression models. From the number of Instagram check-in posts (visitors), nine attractions, including "Kuibishan Geopark," "Penghu Great Bridge," "Qimei Twin-Hearts Stone Weir," "Magong Duxing 10th Village," "Aimen Beach," "Shanshui Beach," "Whale Cave," "Fenggui Blowholes," and "Daguoye Columnar Basalt," maintain high levels of attractiveness both during and outside of the pandemic. Additionally, "Penghu Paradise Road " entered the top ten most attractive spots from March to August 2020, and its characteristics and attractiveness are similar to the top spot, "Kuibishan Geopark," requiring timing with the tides to reveal its mystery, making it a highly potential tourist destination.

Atefeh, Farzindar, and Wael Khreich. 2015. 'A survey of techniques for event detection in twitter', Computational Intelligence, 31: 132-64.
Bao, Jie, Yu Zheng, David Wilkie, and Mohamed Mokbel. 2015. 'Recommendations in location-based social networks: a survey', GeoInformatica, 19: 525-65.
Brian Holak, and Emily McLaughlin. 2017. 'DEFINITION Instagram'.
Carvache-Franco, Orly, Mauricio Carvache-Franco, and Wilmer Carvache-Franco. 2022. 'Coastal and marine topics and destinations during the COVID-19 pandemic in Twitter's tourism hashtags', Tourism and Hospitality Research, 22: 32-41.
Dolan, Rebecca, Jodie Conduit, Catherine Frethey-Bentham, John Fahy, and Steve Goodman. 2019. 'Social media engagement behavior: A framework for engaging customers through social media content', European Journal of Marketing, 53: 2213-43.
Gao, Song, Krzysztof Janowicz, and Helen Couclelis. 2017. 'Extracting urban functional regions from points of interest and human activities on location‐based social networks', Transactions in GIS, 21: 446-67.
Goodchild, Michael F. 2007. 'Citizens as sensors: the world of volunteered geography', GeoJournal, 69: 211-21.
Huang, Haosheng. 2016. 'Context-aware location recommendation using geotagged photos in social media', ISPRS International Journal of Geo-Information, 5: 195.
Huang, Haosheng, Georg Gartner, Jukka M Krisp, Martin Raubal, and Nico Van de Weghe. 2018. 'Location based services: ongoing evolution and research agenda', Journal of Location Based Services, 12: 63-93.
Majeed, Mehwish, Muhammad Irshad, Tasneem Fatima, Jabran Khan, and Muhammad Mubbashar Hassan. 2020. 'Relationship between problematic social media usage and employee depression: A moderated mediation model of mindfulness and fear of COVID-19', Frontiers in Psychology, 11: 557987.
Pérez-Vega, Rodrigo, Babak Taheri, Thomas Farrington, and Kevin O'Gorman. 2018. 'On being attractive, social and visually appealing in social media: The effects of anthropomorphic tourism brands on Facebook fan pages', Tourism management, 66: 339-47.
Pachucki, Christoph, Reinhard Grohs, and Ursula Scholl-Grissemann. 2022. 'Is nothing like before? COVID-19–evoked changes to tourism destination social media communication', Journal of Destination Marketing & Management, 23: 100692.
Rizwan, Muhammad, Wanggen Wan, and Luc Gwiazdzinski. 2020. 'Visualization, spatiotemporal patterns, and directional analysis of urban activities using geolocation data extracted from LBSN', ISPRS International Journal of Geo-Information, 9: 137.
Tasci, Asli DA, and William C Gartner. 2007. 'Destination image and its functional relationships', Journal of travel research, 45: 413-25.
UNWTO. 2023. 'The End of COVID-19-related Travel Restrictions – Summary of findings from the COVID-19-related Travel Restrictions reports': 28.
Wahyuni, Herpita, Eko Priyo Purnomo, and Aqil Teguh Fathani. 2021. 'Social media supports tourism development in the COVID-19 normal era in Bandung', Jurnal Studi Komunikasi, 5: 600-16.
Zheng, Yu. 2011. 'Location-based social networks: Users.' in, Computing with spatial trajectories (Springer).
中華民國內政部. 2021. '澎湖縣'.
王嘉明. 2008. '澎湖國家風景區觀光資源階層體系調查規劃': 98.
林長郁. 2018. '以群眾力量為基礎之空間資訊檢視商圈活動', 觀光與休閒管理期刊, 6: 1-9.
莊哲瑋. 2016. "陽明山國家公園社群臉書打卡資料分析與應用." In 地學研究所地理組, 129. 台北市: 中國文化大學.
郭芳妤, 詹進發, and 許世宏. 2014. '應用志願性地理資訊於社區物種調查之研究', 地理研究: 83-103.
陳良源. 2021. "澎湖觀光產業與環境資源." In 2021 後疫情時代創新地方治理國際學術研討會會議手冊 (含論文集), b1-49. 中國地方自治學會.
陳哲銘. 2013. '利用自願性地理資訊培養學童的地方感: 從苗栗小學生得到的經驗', 地理研究: 75-90.
策略顧問公司, OOSGA. 2023. '台灣社群媒體現況：2023年社群平台發展趨勢、用戶分佈數據'.
黃友岳. 2015. '適地性社群資料分析在犯罪預測之應用', 國立臺灣大學.
趙惠玉, and 林芳儀. 2020. '澎湖地區之旅遊滿意度：三年期調查', 島嶼觀光研究, 13: 1-26.
澎湖縣政府. 2022. '澎湖縣第六期(112-115年)離島綜合建設實施方案'.
澎湖縣政府主計處. 2021. '統計應用分析報告澎湖縣觀光旅遊發展概況'.
澎湖縣政府全球資訊網. 2023. '探索澎湖-自然環境'.
蔡金倉, and 葉莉亭. 2022. '探討澎湖觀光旅遊之服務創新與效益－以海上花火節為例', 品質月刊, 58: 16-23.
賴宗成, 劉喜臨, and 吳菊. 2023. '疫情下澎湖觀光產業復甦的可能途徑與作為', 中國地方自治, 76: 3-18.
簡玉鳳. 2014. '利用適地性服務建構行動空氣品質預警系統'.