DWBI Vienna – Extracting Regional Indicators of Subjective Well-Being from Digital Content Streams
Funded by the Digital Humanism Program of the Vienna Science and Technology Fund (WWTF), the DWBI Vienna Project is a joint research undertaking of the Research Center for New Media Technology and the Department of Sustainability, Governance and Methods of Modul University Vienna. The three-year project will use the semantic technology stack of webLyzard to develop new knowledge extraction algorithms. The goal is to accurately measure well-being in the Vienna region, and to compare the obtained results with those of other selected cities. A visual analytics dashboard will help to better understand patterns and changes in subjective well-being by region and over time.
Quality of Life Metrics for the City of Vienna
Vienna has been voted Best City for Quality of Life for many years. How can stakeholders get insights into perceived well-being in a timely manner? How can we relate these insights to specific aspects of urban environments? Conducting a comprehensive national survey is costly and requires significant organizational effort. Moreover, they need to balance knowledge to be gained and burden placed on respondents. Surveys take a lot of time to compile until official results can be published. Since predefined questions only allow for limited customization, they also tend to lack regional information.
The rather generic nature of certain well-being metrics (e.g. “satisfied with life”) constrain their usefulness to support policy decisions. To address these challenges and provide insights into well-being and related social issues, we propose AI-based opinion mining in conjunction with knowledge graphs to extract affective knowledge from user-generated digital content. Automatically mapped to existing indicators, this knowledge will complement and enrich surveys and traditional assessment methodologies. Such computational approaches to process online data provide a near real-time analysis of evolving citizen perceptions.
Insights from Digital Humanism Research
The concept of Digital Humanism puts people back at the center of technology development and benchmarking efforts. Aligned with this goal, the methods of DBWI will help detect relevant trends in subjective citizen experiences. They will also reflect the level of agreement with specific social or political measures. We will investigate, for example, whether user-generated content from social media reflects topics commonly considered Viennese strengths such as local recreational areas. A pilot study has shown that participants mention local factors spontaneously in face-to-face interviews. We aim to explore whether this is also the case in online statements.
The computations will not only quantify sentiment and standard human emotions, but also domain-specific affective categories reflecting well-being in an urban environment. The system will compute indicators with respect to the relevant named entities, e.g., city or district. We will contrast the results obtained for Vienna with at least one other city in Austria (e.g. Graz) and one in Germany (e.g. Munich). Furthermore, we will investigate district-level deviations. Particular emphasis will lie on local policy-related aspects such as infrastructure, but also more humanistic aspects. This includes the quality of social relationships, for example, social cohesion, treatment of minorities or trust in institutions. In other words, DWBI Vienna will investigate social capital on all societal levels.
Related Research Projects
DWBI Vienna will evaluate trends along given threads of interest, such as policy reactions to COVID-19 and their immediate and long-term consequences. This aspect of the planned work will benefit from the Corona Mood Barometer. The latter stems from the Horizon 2020 Project ReTV and the FFG-funded project EPOCH. Given the outstanding reputation that Vienna enjoys in international Quality of Life (QoL) Rankings, we anticipate that our results to reflect this positive assessment. The added value will consist of a detailed profile showing aspects with varying needs for action, indicating what pleases and bothers Viennese citizens.
To generate trend reports and contrast objective and subjective indicators, we will analyse geographic or demographic determinants of emotional expressions where possible. This will reveal where developments are subjectively under- or overrated, as compared to common assessments based on objectively measured circumstances. We will then identify the reasons leading to such judgement gaps, for example societal misperception. This will include attempts to quantify assessment distortions due to differences between spontaneous and coordinated online publishing. The latter refers to suspicious patterns and manipulation attempts, for example a large number of postings published concurrently via multiple accounts.