Ecostack Innovations Helps Uncover Visitor Preferences for Nature-Based Tourism in Malta

A recent study published in the journal Ecosystems & People by researchers, Costadone and Balzan, has shed new light on visitor preferences for nature-based recreation in Malta. The study employed a novel approach that combined machine learning and image recognition to analyse a vast dataset of social media posts from Flickr and iNaturalist. The findings provide valuable insights into how nature-based recreation varies across the Maltese landscape, particularly during different seasons. 

Utilising Social Media Data for Sustainable Tourism

The researchers extracted a total of 41,277 photos from Flickr and 3,551 observations from iNaturalist, spanning the years 2015 to 2021. Each image was carefully analysed using a machine learning algorithm powered by Google Cloud Vision, which assigned ten labels to each photo based on its content. These labels were then classified into different categories of natural features, such as landforms, animals, and vegetation through a validated machine learning algorithm.

Highlighting Popularity of Coastal Areas and Nature Reserves

An analysis of the labels revealed a high concentration of images depicting nature-based interactions in coastal areas, nature reserves, and the green spaces surrounding Valletta and the Grand Harbour. The researchers observed a higher frequency of social media posts from urban locations during the wet season, while rural areas and landscapes in Gozo and Comino attracted more attention in the dry season.

Connectivity to Road Networks and Socio-Demographic Factors

The study also investigated the relationship between the distribution of social media posts and the road network. The analysis found a strong correlation between road accessibility and the frequency of nature-based recreation activities, emphasizing the importance of infrastructure in driving visitation to these sites.

Moreover, the researchers examined possible socio-demographic factors influencing nature-based recreation patterns. They discovered that both Flickr and iNaturalist platforms registered a high concentration of nature-based activities in areas according to relative social advantage and disadvantage. Both platforms registered a high concentration of nature-based activities in areas with a high percentage of beneficiaries of social welfare aimed at the unemployed and elderly demographic groups, implying that disadvantaged districts are highly important to nature-based visitation.

Harnessing Crowdsourced Data for Conservation Management

The findings of this study highlight the potential of open-access social media data for informing conservation management strategies of natural and protected areas. The use of machine learning algorithms in conjunction with manual image analysis provides a robust, time-efficient, and cost-effective approach to evaluating visitor preferences and identifying areas that require greater attention for sustainable tourism.

By analysing social media posts on platforms like Flickr and iNaturalist, decision-makers can gain valuable insights into the preferences of visitors and make informed decisions to enhance visitor experiences while minimizing environmental impacts. This approach is particularly valuable in densely populated coastal areas, where ecosystems are more vulnerable to degradation.

Interested to learn about our work on nature-based tourism? Our team at Ecostack Innovations, has been working with key stakeholders to analyse nature-based tourism and is using a combination of mapping techniques and analysis and machine learning to analyse visitation and opportunities. Get in touch to learn more about or work or if you'd like to collaborate with us.