Contributions published at Social Computing (Aniko Hannak)

Many users tend to seek information about specific countries on TikTok before planning international travel. To investigate the objectivity and authenticity of national information on TikTok, we examine the national images presented on TikTok for 12 popular tourist countries from different continents. We utilize web scraping techniques to collect TikTok data for these countries, including descriptive data of videos, watermark-free video contents, comments, etc. Subsequently, we conduct separate analyses of descriptions, video contents, and comments of TikTok videos related to each country. Additionally, we design a survey questionnaire to gather user perceptions of various national images on TikTok. Ultimately, through the analysis of TikTok data and survey responses, we identify consistent trends and specific themes in the descriptions, video contents, and comments of TikTok videos related to specific countries. For instance, TikTok videos related to Argentina predominantly revolve around the theme of football, while those related to Italy and Spain primarily focus on travel and food. Furthermore, users' impressions of specific countries on TikTok closely align with the national images presented on the platform.

This thesis explores the influence of external factors on the decision quality of professional football players in shot-taking scenarios. Based on innovative freeze-frame technology, this thesis developed an expected goal model that quantifies decision quality, enabling the analysis of the factors score, time played, competition stage, and market value. The results of this study make a substantial contribution to the field of football analytics, enhancing the understanding of the complexities involved in the dynamics of decision-making. The methodological approach results in the utilization of the expected goal model to evaluate the quality of decisions made during shots in three major tournaments. The outcomes of this investigation reveal correlations between decision quality, the game score, and players’ market value. Notably, a heightened sense of self-confidence, influenced by a favourable game score, reinforces decision-making. Moreover, players with greater market values tend to exhibit superior decision-making skills. However, the study did not yield statistically significant relationships between decision quality and the duration of playtime or the game’s competition stage. Findings offer practical implications for coaches, who can enhance player self-confidence, improving decision-making, and managers who can use market value as an indicator for decision quality. In conclusion, this thesis advances the field of football analytics by researching how external variables impact the quality of decisions during shot-taking scenarios. The study underscores the role of self-confidence and market value as indicators of effective decisions. Furthermore, this research enriches the understanding of the decision-making processes intrinsic to football, thereby offering insights germane to player development and team management. Additionally, the thesis underscores the possibilities of freeze-frame technology and how, even with limited resources, a robust model for quantifying decision quality can be constructed. Future work should expand the model’s scope and examine more possible factors influencing decision quality.

This study investigates recruitment criteria in the Internet and finance industries across key Chinese cities: Beijing, Shanghai, Guangzhou, Shenzhen, and Hangzhou. Utilizing a web crawl technique to get online recruitment platform data (https://www.zhaopin.com/), we examined 174,016 job postings and administered a questionnaire to explore recruitment discrimination. Using Chinese word segmentation and relevant techniques, our findings reveal variations in job opportunities, educational preferences, salaries, and essential skills between the Internet and finance sectors in key Chinese cities. Recruitment discrimination rates fluctuate across cities, with Shenzhen reporting elevated rates. Education discrimination prevails, accompanied by age and gender discrimination. Notably, female individuals are more likely to perceive gender discrimination.

This study examines how different types of users experience and perceive gamification on Stack Overflow. In addition, the game balance and users’ opinions on the level of challenging tasks for members of different skill sets are explored. The study data were obtained using a mixed method approach that combined quantitative and qualitative methods. Quantitative data were collected through a survey, and qualitative data were obtained by interviewing 10 Stack Overflow users. The results suggest that guidelines that are applied in Stack Overflow reduce the need for competence and autonomy, and as a result, discourage expert and novice users from contributing to Stack Overflow. Furthermore, the Stack Overflows’ reward system awards trendy questions over complicated and niche questions. The results indicate that novice users may feel that they cannot contribute to the platform. In addition, they struggle with finding adequately challenging tasks to solve, resulting in them being discouraged from contributing.

The measurement of credibility for Twitter content has gained significant attention due to the difficulty in verifying the accuracy of posts, particularly those made by users who identify themselves as experts by including titles such as Dr.’ or M.D.’ in their display name. This study aimed to investigate three research questions. First, we assessed the credibility of users who display qualified titles on Twitter. Next, we analysed the types of viewers who are most susceptible to the influence of such users, and finally we proposed strategies that can be used by actual ‘Dr.’ titled users to enhance their credibility on the platform. To gather data, a between-subject experiment and a survey were designed and conducted. The results indicate that users with professional titles in their display names are perceived as more credible than those without such titles. Additionally, the study found that individuals who have never used Twitter before are the most impacted by Twitter content. Our study suggests that real ‘Dr.’ titled users can increase their credibility by including a relevant bio in their profile and by including paper links in their tweets. By doing so, these users can more effectively persuade the public of their expertise.

This thesis conducts an explorative data analysis of People also asked (PAA) questions and answers on Google. The study uses web scraping techniques to collect PAA data for various search terms in the domain of complementary and alternative medicine (CAM). By performing an algorithmic audit, we show that inside the US, neither the location nor the search history influences the set of questions and answers a user is presented by Google for a given search term. The collection of PAA data for an array of real-world search terms in the domain of CAM reveals that many of the answers provided by Google are not from independently fact-checked sources, but instead biased websites such as retail businesses or special interest advocacies. Our results further suggest that the question and answer pairs in PAA might lead to confirmation bias.

null

This thesis investigates the change in the amount of travel in the countries France, Sweden and Japan during the COVID-19 pandemic on the example of booking.com. On top of that, the response strategies from France, Sweden and Japan are taken into account and it is evaluated how they influence travel behaviour. Additionally, recurring topics in guest reviews in relation to COVID-19 on booking.com get examined. To examine those issues review data from 535 high rated hotels and residences on booking.com has been collected and analysed. The results show that the amount of travel decreased significantly during the pandemic and that the response strategies of countries influenced the travel behaviour in each of the investigated countries in a different way. Recurring topics in reviews could be narrowed down to a top three, those are 'good safety measures', 'good food experience', 'bad food experience'.

Recent research has shown that the way search results are presented has a major influence on consumer choices and search behavior. The main reason for such choices is that higher-ranked results attract more clicks and attention compared to lower-ranked results. The introduction of featured snippets resembles a substantial change to the overall appearance of the search engine result page and provides a short answer to user queries. This thesis presents a customizable search behavior tracking tool and a controlled web-based experiment on featured snippets using said tool. The goal of the experiment is to observe if and how the presence of a featured snippet affects search behavior, the success of the search, and the confidence in the found answer. The results of the experiment show, that (i) clicks and total search time decrease significantly when a featured snippet is present, (ii) featured snippets that contain a complete and correct answer significantly increase the likelihood of finding the correct answer, and the participants’ confidence in the found answer, and (iii) featured snippets with a partial answer show no significant increase or decrease in the likelihood of finding the correct answer or the participants’ confidence in the found answer. Thus the findings point to the fact that featured snippets on the one hand save users’ time in finding answers, but on the other hand, may also discourage them from taking a variety of sources into consideration. Resultingly, it could be theorized that the implementation of featured snippets could overall lead to less informed decisions.

This thesis examines the public value creation in digital ecosystems. It, therefore, aims to elaborate on success factors that lead to successful public value creation in digital ecosystems. For this purpose, interviews were conducted with representatives of digital ecosystems as part of a multiple case study, which was then evaluated using qualitative analysis. Through the analysis combined with the literature review, a number of success factors were determined that positively influence the process of public value creation. In a second step, a fuzzy-set Qualitative Comparative Analysis (fsQCA) was performed, which compiles configuration paths from these success factors. The results of this configuration analysis suggest different configurations paths that show in which configurations the success factors lead to successful public value creation for platform owners and complementors.

Ride-hailing services such as Uber and Lyft have become globally pervasive in the last decade, revolutionizing the taxi sector for both customers and drivers. This business model breaks many barriers of entry for new drivers and makes commuting by taxi cheaper and more convenient for customers. Nevertheless, it is also affected by drastic income inequalities and weak job security. The goal of this thesis is to investigate whether we can reduce income inequalities by using ML-based demand prediction to strategically dispatch drivers. As such, I first develop a machine learning model to predict customer demand using a real-world ride-hailing dataset collected by the city of Chicago. Secondly, I integrate the real-world data within an agent-based model to make an initial exploration of the potential of using the predictions for decreasing the income inequality of drivers. The results show that (a) the prediction model is able to fairly accurately predict demand, and (b) one of the three implemented rule-based naive dispatchers successfully used the predictions in order to increase the level of fairness.

Research about the acquisition of expertise provides insights into the mental processes of the human mind and allows us to improve our learning methods. Cognition researchers often perform experiments with standardized tasks which allow them to observe and measure these mental processes. The game of chess proved to be of high value for such explorative experiments, leading to findings of pattern recognition, memory capacity, and problem-solving strategies. Our work provides empirical evidence about the importance of opening familiarity in relation to general calculation abilities in chess. We recruited 297 chess players via social networks who solved 32 purposefully selected chess tasks each in an online setting. Contrary to previous studies, which focused primarily on expert chess players, we study the tactical problem-solving abilities among amateur-level players. Our results show that opening selection significantly shapes the tactical pattern recognition of beginners in the early stages of chess development. This effect diminishes with an increased skill level to a point at which puzzles are solved equally well regardless of opening familiarity. These findings are in line with established theories of skill acquisition in chess.

null

This bachelor thesis will look at hotel ratings and whether these reviews have similarities in the rating (score) concerning demographic aspects, such as whether a person has traveled alone or in company or from which country this person is from. In this thesis, the focus is placed on Booking.com and Ibis Hotels. Explanations for these differences are also sought, such as whether Booking.com offers different prices for different people. To determine whether there is a link between demographic aspects and the score, over 1.5 million reviews from Booking.com were scraped and analyzed using regression.

TikTok is the fastest growing social media platform of the 21st century with over 1.5 billion active users of which the majority is from generation Z. Its most important success-driver is its recommendation system. With its impressive growth, TikTok has attracted many researchers, but little work has been done on finding empirical evidence for the influence of certain factors on the recommender system deployed by TikTok. Our work lays the foundation to fill this research gap. We have developed our own scraping algorithm with which we tested and analysed the effect of the language and location used to access TikTok, follow- and like-feature, as well as how the recommended content changes as a user watches certain posts longer than others. We provide evidence that all of our tested factors influence the content recommended to TikTok users. Further, we identified that the follow-feature is the strongest influential factor, followed by the like-feature and video view rate. Our analysis reveals the creation of filter bubbles and weighting the video view rate as too high may result in severe consequences, specifically considering naive and gullible people, such as children, who are easy to manipulate.

The success of the gig economy has led to workers whose employment status is often defined by short-term commitments. Companies like Uber promote flexible side-job opportunities, but their drivers are frequently underpaid and uninsured. Uber drivers have fewer tools to communicate with each other and are only now seeing political support of organizations. Twitter poses an opportunity for drivers to join forces, communicate effectively, and create a collective identity. In this thesis, I collect and analyze Twitter data related to Uber's IPO and the strikes preceding it. Using a combination of PCA, k-means, and NLP techniques, I characterize different communities. I identify patterns in discussed topics, hashtag usage related to real-life events, and between expressed sentiment and the number of retweets. Discussions of Uber-related cultural conflicts can be observed in high-emotion tweets. The goal of this work is to complement existing branches of sociological, legal, and algorithmic fairness literature.

Stack Overflow (SO) is recognized as a technical knowledge-sharing market where goods and services are merely based on asking questions and providing answers. The majority of the questions are related to technology and coding problems. Each year SO publishes a survey with an idea to reach out to coders across the world and to gain more insight into its users and their experience on the platform. SO does everything to serve the needs within the developers' community. The data obtained from the annual survey helps to make changes and set goals to improve the environment and make it more welcoming and inclusive of the SO community. The community does only include visitors to SO, but also everyone who codes or does some coding in their work or studies. The survey questionnaire starts with questions about user demographics, coding interests, current company experience, preferences for different coding languages, and getting feedback on leading technologies of the time. To maximize the accuracy of results, the platform has a minimum threshold for total time spent by each candidate in completing the survey. The platform provides a census badge to its users after completing the survey. The badge falls under the silver badge category and exhibits high reputation scores. This study is a quantitative attempt to understand the differences among the users who participate in the developers' survey to the ones who do not participate. We wanted to identify the key factors that may influence participation in the survey to gain better understanding of the population that takes the survey and how they differ - if at all - from the rest of the community. It also aimed to help us understand the attitude of underrepresented groups such as women and non-active users towards the developers' survey. Our findings suggested that the majority of survey respondents belonged to the community of users with high reputation scores on the website. The users with high tenure on the website were also more likely to participate in the survey. The self-promoters - users who actively promote themselves on the website, and also on other social media platforms such as LinkedIn, Twitter, and Github - were among the majority of survey participants. In terms of user activity on the website, 85% of the survey participants were active answer providers. We also aggregated the participation from the level of participation of the users to the level of geographical regions and learned that the users from the continent of Oceania were the principal contributors in the survey, followed closely by those from Africa and South America; Europe came a distant fourth, and the contribution rate of users from Asia was the lowest of all. We could not find statistically signicant results for users from North America. It is inquisitive for us see if the census badge leads to more participation. Our study suggests that in 2017, 40% of the respondents claimed the badge, followed by 50% in 2018 and 60% in 2019. The participation-to-badge-claim ratio has increased by 10% each year from 2017 to 2019.

null

null