Research and analysis

Capturing Engagement Numbers - Strand 2: Case Study: Giant's Causeway

Published 13 March 2026

This report was authored by Jack Medlock, Hannah M. P. Stock, Andrew Knight, Donna Phillips, Adam L. Ozer, and Joseph Stordy at Verian, Dr Michael Sinclair, Dr Craig Macdonald, and Prof Iadh Ounis at The University of Glasgow, and Faculty.

This research was supported by the R&D Science and Analysis Programme at the Department for Culture, Media & Sport (DCMS). It was developed and produced according to the research team’s hypotheses and methods between October 2023 and June 2025. Any primary research, subsequent findings or recommendations do not represent UK Government views or policy.

Executive Summary

While ticketing data gives a good understanding of engagement with ticketed events, measuring engagement at un-ticketed events is difficult and often relies on surveys. Although the data provided by these surveys is of good quality, replicable and can provide demographic insight they come with their own challenges and limitations. They are limited, for example, in their ability to comprehensively measure local participation at specific events or spaces. Given even ticket sales or traditional crowd counting methods may not be an accurate reflection of attendance numbers, DCMS want to explore new data-driven methods using novel techniques.

This case study explains and compares 3 such methods, each based around a specific data source, for predicting attendance at the Giant’s Causeway; Aerial photography data (Strava), Mobile app data (Huq) and Social media data (Pulsar).

Each methodology is summarised and compared against 8 categories: Accuracy, Bias, Ethics, Deliverability, Cost, Demographics, Generalisability and Accessibility.

Case studies have been developed for each event in scope of the research The British Museum was one of 5 events selected.

1. Event Overview

Location: The Giant’s Causeway, Northern Ireland, UK
Summary: Estimating visits over the course of one calendar year.

The Giant’s Causeway was selected as a research candidate for the counting engagement project because it exhibited the following characteristics:

• One of the most visited outdoor cultural sites in the UK each year events in the UK, with 660,000 visitors in 2023.

• High density of moving targets, with lots of churn over time makes counting attendance more challenging.

• Complexity of background geography presents interesting challenges for image-based crowd counting with object detection machine learning models.

• Availability of open-sourced drone video footage from the internet to test image-based modelling approaches.

• International fame of site means online posts about the Causeway are mostly not from people directly attending the location, with creates interesting challenges for predicting attendance from social media posts.

2. Methods

Three methodologies were developed to predict attendance at the Bradford City of Culture

Aerial Photography
Running Machine Learning (ML) models on high-vantage aerial footage of crowds, typically captured by CCTV cameras or drones, can be used to estimate crowd sizes.

Object detection and crowd density estimation models can analyse this footage to identify and count individuals within the captured area. These models are trained on large datasets of annotated images, enabling them to perform well even in dense and complex crowd scenarios. They do not use facial recognition.

Video footage of the Giant’s Causeway taken from drones and licensed online was experimented with, using a range of machine learning models to count crowds attending the location.

Mobile App Model
Many applications collect real-time location data on mobile phones or GPS-enabled devices. These data points are routinely anonymised and sold to third parties(detailed in the T&C’s when you sign up to the app) which collate and aggregate this information to provide population level estimates of people’s locations.

Data providers (such as Huq) create dashboards, based on a raw feed of real-time data. These data dashboards allow users to visualise data at different points of interest, defined by user-generated boundaries.

Aggregated underlying data from Huq was accessed to analyse the outputs for specific boundaries of the Giant’s Causeway and estimate attendance.

Social Media Data
User-generated content posted on social media platforms can contain information about people’s locations. These digital traces – such as status posts and photos and any public information in their profile about location - can be used as a proxy for physical attendance.

Social media platforms have commercial agreements with social media monitoring companies, such as Pulsar (who we used for this work), who provide a service allowing users to search for and collect posts fully in compliance with the platform’s terms and conditions.

Anonymised social data from Pulsar was analysed and a classifier model was then used to predict overall attendance using posts that can be determined as from true attendees of the Giant’s Causeway.

3. Mobile App Data Methodology

There are a six key steps (covered in 3.1 - 3.6) in the process of measuring site and event visitation using mobile phone data:

1. Defining Boundaries

2. Extracting Mobile Phone Data

3. Collecting Ground Truth Data

4. Site Feature Extraction and Engineering

5. Scaling and Estimating Visits using Sample Weighting

6. Modelling Visition

3.1 Defining Boundaries

Summary: Establishing the geographic boundaries of the site or event for data collection. Ensuring accurate delineation to minimize data spillover from surrounding areas.

Figure 1: Established geographic boundary for The Giant’s Causeway

Boundary Available: Where the site boundary is available through Open Street Maps, it is obtained by querying the Application Programming Interface.

Boundary Unavailable: Where the site boundary is not available from Open Street Maps, it is drawn manually using Geographic Information Systems software based on the Open Street Maps base layer.

3.2 Extracting Mobile Phone Data

Summary: Extracted geographic boundaries serve as the foundation for collecting geospatial mobile phone data, capturing GPS-recorded device locations within the defined area.

Figure 2: Capturing GPS-recorded device locations within the defined area

Collected Data for site of interest:

• Unique Mobile Users – Number of distinct devices detected at the site.

• Number of Mobile Visitor Days – Total device visits over a given period.

• Spatial Patterns of Use – Sensitivity of GPS data within the site.

• Visitor Catchment Areas – Geographic origins of visitors to the site.

• Estimated Geo-Demographics – Socio-economic characteristics inferred from visitors’ home area.

• National Mobile User Panel – Broader panel dataset used to weight and scale mobile visitation to reflect the total population.

3.3 Collecting Ground Truth Data for Modelling

Summary: Baseline data is collected from a diverse range of sites across the UK to support the modeling approach. This data provides ground truth annual visitor numbers, allowing models to learn patterns based on mobile data and site characteristics.

Figure 3: Open Street Map boundaries for visitor attractions for AVLA (Association of Leading Vistor Attractions) sites.

ALVA Annual Visitation Data Collected for (2019-2023); 251 Tourist sites Across UK, 928 Observation. Examples: Carisbrooke Castle, Imperial War Museum, Tate Modern, Pollok House.

3.4 Site Feature Extraction and Engineering

Summary: For use in the modelling approach, various site and context specific characteristics were collected or calculated for each location in the ground truth data. These features provide additional context and help explain baseline visitation patterns.

Figure 4: Site features inputted to create visitation estimates

Figure 5: These features provide additional context and help explain baseline visitation patterns.

3.5 Scaling and Estimating Visits using Sampling Weighting

Summary: The mobile phone population represents only a subset of total visitors to a site, so it is necessary to weight and scale the data to account for under- or over-representation across areas and produce population-level estimate of visitation.

Figure 6: Graphical depiction of sample weighting visitation estimates

Each mobile user is assigned a home area monthly. To estimate total visitation to the site, visitors are weighted by their home County to correct for over- or under-representation of mobile users across the UK. Visits are then scaled to a population-level estimate using the ratio of mobile users to the adult population in each County. This process is applied daily across all Counties with recorded visitors, and the final estimate is the aggregation across Counties.

3.6 Modelling Visitation

Summary: Using the baseline data as the dependent variable, various aspects of mobile data are then incorporated, including sample-weighted visitation estimates and site-specific features, to train and test a range of models. These are deployed to estimate visitation at the site.

Figure 7: Flowchart depicting how visitation predictions are created

3.7 Results of Mobile App Data Methodology

Mean Absolute Error: 20,503 : The average difference between the actual attendance and the attendance our model predicted, measured in the same units as the data (in this case number of people). It shows, on average, how many people our predictions were off by. A smaller number means the model is better at making accurate predictions.

Percentage Error: 9.5% : The average difference between the actual and predicted attendance, expressed as a percentage of the actual attendance. This tells us the size of the error in relative terms and is helpful for understanding how big the error is compared to the event’s size. A lower percentage means the model is performing well.

Predicted Attendance: 2.7M (Actual 2.87M): The number of people the model thinks attended the event.

These results show the median values of estimates across 4 years (2020-2023). The median is used rather than the mean to reduce the impact of the Covid-19 pandemic distorting the results.

3.8 Limitations

There are a few key limitations with this approach of leveraging location data to predict overall attendance at events:

• Boundary Specification: Given the geospatial nature of the data, the process of defining the boundary of a site directly affects the extracted data and, consequently, the visitation estimates. Inaccurate or inconsistent boundary delineations may lead to underestimation or overestimation of visitor numbers, particularly for sites that lack clearly defined perimeters or those with complex spatial layouts.

• Accuracy and Impact of Surrounding Areas: Mobile phone data inherently contains positional inaccuracies due to a range of factors. This can result in data from adjacent areas being incorrectly attributed to a site, particularly when the site is surrounded by roads, transit hubs, or densely populated urban infrastructure. Such spillover effects may introduce systematic biases in visitor estimates.

• Time Period and Data Volume Constraints: The choice of the study period and the availability of data within specific time windows impact the reliability of visitor estimates. Shorter time periods may result in lower data volumes, leading to increased variability and reduced confidence in the estimates. This limitation is particularly pronounced in less frequently visited sites or during off-peak periods, where mobile data penetration may be lower. Additionally, fluctuations in data availability across different seasons, days of the week, or special event periods can introduce inconsistencies in trend analysis.

• Baseline Ground Truth Data Availability:

  • Volume and Diversity of Sites: The reliability of any model built on mobile phone data is contingent on the availability of high-quality baseline ground truth data. If the model is trained on a limited or non-representative set of cultural sites, extrapolating results to different types of locations (e.g., from museums to outdoor heritage sites) may introduce inaccuracies due to differences in visitation patterns

  • Time Period of Baseline Data: The validity of model outputs is constrained by the time periods for which ground truth data is available. If reference datasets do not cover short-term variations or seasonal fluctuations, the model may struggle to produce accurate estimates at finer temporal scales, such as daily or hourly visitation patterns.

• User Demographics and Sample Bias: While some research has shown that mobile phone data provides a good fit to the general population in terms of geographic and socio-demographic coverage, it still represents only a small percentage of the total population. When analysing smaller spaces or shorter time periods, the subset of available data is reduced further, increasing the potential for sample bias in the output.

4. Social Media Methodology

There are 4 steps (covered in Sections 4.1 - 4.4) in the process of estimating attendance from Social Media Methodology:

1. Data Collection

2. Data Extraction

3. Data Processing & Classification

4. Scaling Model

4.1 Data Collection

Many social media companies have strict policies governing the use of their APIs. To ensure compliance with these policies the Pulsar platform was used to access social media posts at scale. Pulsar provides an API and managed wrap-around service that allows users to collect and analyse social media posts across multiple sources in compliance with the terms of service on the relevant platforms. This provided the primary data set for our social media analysis.

To protect user privacy, all data gathered across the platforms explored – Facebook, Instagram, X, Trip Advisor, Reddit - was anonymised by replacing usernames with randomly generated IDs.

To structure the data collection effectively, two main types of queries were employed on the Pulsar platform:​

Live Queries​​: Designed for ongoing events, this was the type of query used for the British Museum, these queries collect data over a defined period. The primary data sources for these queries were Instagram (limited to live data only) and Facebook (limited data to the past 30 days only - set by Meta’s Terms and Conditions).

Historic Queries:​​ Used for past, one-off events, such as the London Marathon, capturing data from one week before and after the event. The key data sources include Twitter, TripAdvisor, and Reddit. Having access to both live and historic data means modelling approaches using social media data are more flexible and can be used for a wider range of events.

4.2 Data Extraction

Searches for social media posts that can be used to predict attendance are constructed on the Pulsar platform using a Boolean search query defined by the user and then refined for each platform’s specific requirements.

Our methodology used large language models (LLMs) to generate queries, albeit with manual oversight. Additional adjustments were also made using event-specific details to improve accuracy and relevance. The Pulsar platform offers a Boolean generator support tool to support users with constructing the validated syntax.

Once a search is running on the Pulsar platform, whether live or historic, it is assigned a unique search ID. This ID, along with specified start and end dates, is then used within code notebooks to download the social media data that has been pulled into the platform.

4.3 Data Processing & Classification

• Classification of Posts for Event Attendance: After downloading the data, each social media post must be classified to determine whether the author attended, plans to attend, or is likely to attend the event. This classification is performed using a Large Language Model, which analyses the content and context of each post. This project used a Llama model, which can be run locally on an inexpensive GPU.

• LLM Query Construction: To accurately classify posts, a structured query is created for the LLM. This query includes:​

  • General Instruction: The LLM is prompted with a directive such as; “You are a helpful assistant for classifying posts about event attendance.”​

  • Event Details: The event name and description are provided. If necessary, an external source can be used to obtain a more detailed event description (e.g from the event website).​

  • Example Posts: To improve accuracy, the LLM is supplied with sample posts that illustrate different classifications. These examples help it distinguish between; users who attended the event, users who intend to attend the event, users who only engaged with the event remotely (e.g., watching on TV).

• LLM Output and Post-Classification: The LLM analyses each post based on the provided query and assigns a classification. It outputs; “1” if the post indicates that the user attended the event, “0” if the post suggests the user did not attend. This classification allows for structured data analysis, providing insights into event participation trends based on social media activity.

4.4 Scaling Model

Purpose The primary objective of the scaling model is to estimate overall event attendance from the number of people posting on social media about being at the event. Since social media posts only represent a subset of actual attendees, once the model has classified social media posts correctly into attendees and non-attendees, it needs to extrapolate this figure to make an estimate for all attendees at the event.

Figure 8: Flowchart depicting how visitation predictions are created

Process

• Input: The model takes as input the number of “positive posts” (e.g. attending the event) classified by the LLM. “Positive posts” refer to social media posts made by individuals who are likely to have attended the event.

• Extropolation: The scaling model extrapolates from the number of positive posts to estimate the total attendance.

• Trained Model and Preprocessing: Firstly, the model will standardise the data collected from different social media platforms so they can be processed within the same model and remove outliers like posts with limited content (e.g. posts solely with emojis). Based on a series of features (a sample of the most important listed below) the model then predicts attendance values based on these structured inputs; Event type (Categorical feature indicating event type), Log_Pulsar_attendance (Logarithmically transformed attendance from the social media posts), Engagement rate (Prevalence of likes + shares + comments), Sentiment norm (Sentiment calculated with Pulsar sentiment analysis tool).

• Ouput: The scaling model produces an estimate of event attendance; For historical events (e.g., concerts, rallies), the model provides a total attendance estimate based on data collected one week before to one week after the event; For ongoing events (e.g., British Museum), the model predicts weekly attendance for a two-week period.

Figure 9: Graph shows model prediction of attendance at a sample of test events used to train the model (y-axis) against the baseline attendance for events (x-axis). The closer to blue dots are to the dashed red-line, the more accurate the prediction.

4.5 Results of Social Media Data Methodology

• Absolute Error: 2380 : The average difference between the actual attendance and the attendance our model predicted, measured in the same units as the data (in this case number of people). It shows, on average, how many people our predictions were off by. A smaller number means the model is better at making accurate predictions. It should be noted the baseline figure of 4000 is itself a rough estimate, so it may be our model was more accurate than the baseline.

• Percentage Error: 56.1% : The average difference between the actual and predicted attendance, expressed as a percentage of the actual attendance. This tells us the size of the error in relative terms and is helpful for understanding how big the error is compared to the event’s size. A lower percentage means the model is performing well.

• Predicted Attendance: 6245 : The number of people the model thinks attended the event.

4.6 Limitations

There are a few key limitations with this approach of leveraging social media data to predict overall attendance at events:

• Query Design: People posting about events on social media will use different language or hashtags to describe the same event, meaning that queries may not collect all relevant social media posts. And additional complication is that queries which are too broad will collect too much irrelevant data (e.g. a query just with the word football will collect billions of posts), which a) can’t be processed and b) lowers performance of the modelling. The main mitigation is using specific, standardised and logical query terms based on a search through social media for how most people are referring to the event. More detail on how to construct these are in the toolkit accompanying this report.

• Platform-Specific Search Methods: Different social media sources require different handling approaches.

  • Instagram: Searches are limited to hashtags, restricting the possible breadth of data collection.
  • Facebook: Keyword-based searches can be more restrictive than fully Boolean queries (e.g., lacking support for nested “AND’s” or “OR’s”).

• LLM Prompting: The classification model used for this research project relies on a relatively small LLM hosted locally to ensure compliance with data governance regulations – in this case not sharing personal data with a third-party (i.e. the model provider). As a result, the locally hosted models used for this research are likely to be poorer performing than larger cloud-native LLMs, resulting in lower accuracy and less reliable estimates for attendance.

• User Demographics and Social Media Habits: The likelihood of event attendees posting varies by demographic and social media platforms, meaning specific demographics may be over or under represented into the model estimates, resulting in both biased results and poorer model performance. Given access to all the demographic data on social media users and the demographics of people who attended this event are not available, there’s no way to correct for this bias within this approach.

• Model drift: As social media habits change over time, the model will be prone to ‘drift’ where the performance degrades, and re-training the model is necessary to ensure continued good model performance.

• Scaling Model and Small Development Dataset: Due to the comparatively small dataset used in this research (e.g. only tens of thousands of social media posts, compared to large datasets of millions of posts), the training of the scaling model was sensitive to outliers and risks overfitting (where the model learns the relationships within the training data but then can’t generalise these learnings to other events). This affected the model’s accuracy and generalisability to some ‘outlying’ events, for example…​

  • Events where ‘Virtual’ Attendance is Possible: If it is available to view on TV or online, there will be more posts with a significantly smaller proportion of people who attended.​

  • Events with Unusual Post Sentiment and Engagement: For example, the Women’s Euro 2022 Final Screenings will have had significantly more posts, with more positive sentiment since England won!​

  • Estimates of Ground Truth: The baseline attendance of events we used for training and evaluating our models were frequently estimates themselves, meaning the final assessment of our results may not be completely fair or accurate.​

• Data Source considerations and restrictions: Due to the platforms T&C’s and behaviour of users on social media platforms):

  • Instagram: Only live data can be collected, restricting access to historical content and is confined to public content.

  • Facebook: Data is limited to the past 30 days and is confined to public content, meaning no private groups can be captured.

  • X: Changes in user behavior have shifted the platform’s use in recent years. Posts indicated a focus on news and commentary, rather than event-related posts, based on a review of collected posts.

  • Reddit: Reddit was found to be a less reliable source for event-related posts, as users are were consistently less likely to share event details on the platform based on our assessment of posts used in the model training.

5. Aerial photography methodology

There are a three key steps (Covered in section 5.1 - 5.3) in the process of measuring event visitation using Aerial Photography methodology:

1. Downloading & Extracting Footage

2. Applying Modelling Approaches

3. Frame Aggregation & Feature Tracking

5.1 Downloading & Extracting Footage

Figure 10: 14 individual scenes extracted from the location

14 Individual scenes were extracted from the location. These were taken from drone footage taken and shared on online that allow licensing for a small fee. From each of these scenes, individual frames were extracted – roughly 4 to 5 frames per second from the original 50 frames per second. This process transformed the video into a series of still images, making it more suitable for detailed analysis and exploitation by machine learning models. By selectively reducing the frame rate, redundant frames were filtered out while preserving those that best captured movement, changes, and key moments within each scene.

5.2 Applying Modelling Approaches

Figure 11: Yolocrowd model output. An individual is counted in a green square, however the model struggled to distinguish between human beings at the location and the hexagonal rock formations that comprise the causeway.

After extracting the individual frames for analysis, a Yolocrowd (You Only Look Once) Model was applied. This is a small-scale object detection model that has been fine-tuned for detecting people in crowds.

However, in the case of the Giant’s Causeway, the model struggled to distinguish between human beings at the location and the hexagonal rock formations that comprise the causeway. This is shown left, where individual rocks are highlighted green, indicating the model incorrectly detecting attendees in these locations.

Overall, the performance of the object detection model was poor when applied to Giant’s Causeway footage, failing to arrive at a plausible count for attendance in this setting.

The model is overcounting attendance because the hexagonal rock formations share visual features with human figures from a top-down drone perspective. Object detection models rely on patterns like shape, texture, and contrast to recognise objects. Here the rocks and humans have similar contours, shadows, and colours, the model may misclassify the rocks as people. This happens because the training data may not include enough examples of the Giant’s Causeway or similar environments, leading the model to generalise incorrectly.

5.3 Frame Aggregation & Feature Tracking

Figure 12: Polygons created around previously identified individuals.

After applying the modelling approaches to the individual frames, the analysis aggregated counts for each scene. To do this, the video was divided into 14 scenes, each containing multiple frames. A SIFT model (scale invariant feature transformation) identified consistent features across these frames, even when the perspective in a frame changes.

Using these features, polygons were created around previously identified’ individuals’, preventing double counting as the analysis progressed through the frames. This approach is designed to arrive at a more total count for each scene by tracking individuals across all frames, rather than relying on isolated snapshots. The analysis also accounts for ‘churn’ or the movement of people into and out of the event area. However, it was assumed that the number of people leaving and entering a scene would be roughly equal, making any impact on the final count negligible.

Again, due to the lack of ‘causeway-like’ images in the training data, the model overcounted attendance, failing to produce a plausible result.

5.4 Results of aerial photography data methodology

Absolute Error: No Result : The average difference between the actual attendance and the attendance our model predicted, measured in the same units as the data (in this case number of people). It shows, on average, how many people our predictions were off by. A smaller number means the model is better at making accurate predictions.

Percentage Error: No Result : The average difference between the actual and predicted attendance, expressed as a percentage of the actual attendance. This tells us the size of the error in relative terms and is helpful for understanding how big the error is compared to the event’s size. A lower percentage means the model is performing well.

Predicted Attendance: No Result : The number of people the model thinks attended the event.

A breakdown of results is not included for the Giant’s Causeway as the inability of the Yolo model to distinguish between people and rocks resulted in extreme overcounting.

6. Comparison of Methods

Below is an assessment of the performance of each methodology in its specific application to the Giant’s Causeway.

6.1 Aerial photography model (Not Recommended): No Results

Explanation

• The object detection model performed poorly, failing to arrive at a plausible count. It overcounted attendance because the rock formations share visual features with humans from a top-down drone perspective. Object detection models rely on patterns like shape, texture, and contrast to recognise objects. Here the rocks and humans have similar contours, shadows, and colours, causing he model may misclassify the rocks as people.

• This happens because the training data lacks examples like the Giant’s Causeway or similar environments, leading the model to generalise incorrectly.

Precaution for use

• This approach should not be applied to the Giant’s Causeway, without first training the model on considerably more data that will improve its performance on similar settings.

Stronger for:

• Accurate results where quality footage is available and approximate figures for churn between frames can be established to reduce double-counting.

• Approach is highly replicable and low cost.

• Innate risk of personal identification in images is unlikely if no facial recognition software is used, as in this project.

Weaker for:

• Model cannot detect people against complex, shadowy or unusual backgrounds unrepresented in training data, as in the case of the Giant’s Causeway

• Access to high-resolution, high-vantage aerial photography is hard to access, requiring bespoke deployment of drones or cameras to events or close collaboration with organisers.

• No additional information is provided about the demographic make-up of crowds.

6.2 Mobile App Model (Recommended): 9.5% error compared to baseline

Explanation

This approach performed well for three main reasons:

• The Giant’s Causeway is a destination site rather than a pass-through location, making their mobile location data a strong indicator of actual attendance;

• Given visitors often spend a reasonable amount of time exploring the site, their mobile signals remain in the area long enough to be reliably captured by data providers;

• Despite the rural location, mobile data signal was still strong enough to support smart phone use.

Precaution for use

• The shortest available collection period for this methodology is 24 hours, meaning this would not be able to predict how many people attended the Causeway over a given afternoon, for example.

• Not all visitors use mobile apps that collect location data, and certain demographics (e.g., older visitors to the Causeway) may be underrepresented, leading to potential bias.

• There are also some ethical concerns associated with the fact that users of these mobile apps do not actively opt in to their data being used to track their attendance.

Stronger for:

• Accurate for long-running recurring events that repeat over the course of weeks, months or years e.g. exhibits or parks. The Giant’s Causeway is a perfect candidate for this.

• Method is low-cost and straightforward to deliver with the right expertise and can be compared to known population distributions for specific sites or events, helping to uncover demographic info and reduce bias.

Weaker for:

• Huq data is generally less suitable for short lived events (e.g. a few hours in duration). It is better suited for recurring events and locations with fixed boundaries.

• Mobile phone location data is subject to some ethical concerns around use. While this data is legal to collect and for providers like Huq to license, there are some concerns about whether individuals can be said have given ‘informed consent’ for their data to be collected.

6.3 Social Media Data (Not Recommended): 235% error compared to baseline

Explanation

This approach performed poorly for the Giant’s Causeway. Possible explanations include:

• The Causeway is a globally famous landmark, widely discussed in the context of NI tourism and folklore. Online conversation from people not visiting in person likely led to the model overestimating attendance;

• Visitors may not post about their visit until much later, making it harder for the model to label social media activity with real-world attendance, compared to passive events in urban settings that attract more real-time posts.

Precaution for use

• Likely to need a significantly larger training data set to accurately label attendance at the Giant’s causeway – a case where real-world attendance and other online posts proved particularly hard to distinguish.

• There are moments when this distinction is likely to be particularly hard to draw. For example, if the Causeway is prominently featured in films, TV shows, or major news stories, online discussions could surge without a corresponding increase in physical visits, further skewing the model’s predictions.

Stronger for:

• Provides a low-cost and straightforward way of leveraging social media data to predict attendance. Pulsar’s managed-API service is intuitive and reduces friction of managing platform-level terms of service obligations.

• Most accurate for events where events have an in-person attendance with significant numbers of posts confirming in person attendance.

Weaker for:

• Accuracy reduced where location is particularly famous and is the subject of many posts from people who have not actually visited the site.

• Incorporation of traditional survey methods into the modelling approach would improve results.

• While compliant with platform-level terms of service, there is not clearly ‘informed consent’ for this use case from an ethical point of view.

• Risks with increased privacy walls means the approach may not be extendable in the future.