Using social media, machine learning and natural language processing to map multiple recreational beneficiaries

Abstract

Information and numbers on the use and appreciation of nature are valuable information for protected area (PA) managers. A promising direction is the utilisation of social media, such as the photosharing website Flickr. Here we demonstrate a novel approach, borrowing techniques from machine learning (image analysis), natural language processing (Latent Semantic Analysis (LSA)) and self-organising maps (SOM), to collect and interpret> 20,000 photos from the Camargue region in Southern France. From the perspective of Cultural Ecosystem Services (CES), we assessed the relationship between the use of the Camargue delta and the presence of natural elements by consulting local managers. Clustering algorithms applied to results of the LSA data revealed six distinct user groups, which included those interested in nature, ornithology, religious pilgrimage, general tourists and aviation enthusiasts. For each group, we produced high-resolution spatial and seasonal maps, which matched known recreational attractions and annual festivals in the Camargue. The accuracy of the group identification, and the spatial and temporal patterns of photo activity, in the Camargue delta were evaluated by local managers of the Camargue regional park. This study demonstrates how PA managers can harness socialmedia to monitor recreation and improve their management decision making

On How Crowdsourced Data and Landscape Organisation Metrics Can Facilitate the Mapping of Cultural Ecosystem Services: An Estonian Case Study

“…Our results continue the methodological approach that has been initiated in previous research [30][31][32]. The LDA topic modelling algorithm significantly facilitated the process of LBSM data assigned with the content-related tags as a result of automated image recognition with Clarifai.…”

“…Therefore, image recognition services and machine learning models have been gaining attention more recently. For instance, machine learning algorithms provided by Clarifai (Clarifai Inc., New York, NY, USA) and Google Cloud Vision were recently reported to be very promising for CES recognition and mapping [30,31], and natural language processing was applied to categorise social media users in relation to outdoor recreation [32].…”

AI Naturalists Might Hold the Key to Unlocking Biodiversity Data in Social Media Imagery

“…This approach could therefore be a tool in the domain of “conservation culturomics,” 38 which uses quantitative analyses to explore changes in human behavior in conservation science. These preferences have been used previously to map multiple recreational beneficiaries, 39 detect human activity patterns, 40 or to quantify the attractiveness of outdoor areas. 41 , 42 Trends in submitted images across years could also be of interest; for example, these could indicate changing levels of interest in wilder forms of gardening or park maintenance that are likely to be of interest to conservationists or those quantifying ecosystems services; photographs of plant-pollinator interactions could illustrate trends in public interest in potential insect declines; increases in images of non-native species could indicate increased awareness of invasive non-native species.…”

Exploring human–nature interactions in national parks with social media photographs and computer vision

“…State-ofthe art computer-vision methods allow for information to be extracted from large volumes of photographs by classifying the content into predefined classes (such as landscapes), by recognizing discrete objects (such as species), or by grouping together similar images for human analysts. These approaches have recently been used to monitor species (Sharma et al 2018) and to examine aesthetic preferences (Seresinhe et al 2017(Seresinhe et al , 2018 and human activities and preferences (Richards & Tunçer 2018;Gosal et al 2019;Koylu et al 2019).…”