Linköping University | Department of Computer and Information Science Bachelor’s thesis, 16 ECTS | Link Usage 2020 | LIU-IDA/LITH-EX-G--20/001--SE Longitudinal study of links, linkshorteners, and Bitly usage on Twitter Longitudinella mätningar av länkar, länkförkortare och Bitly an- vänding på Twitter Mathilda Moström Alexander Edberg Supervisor : Niklas Carlsson Examiner : Marcus Bendtsen Linköpings universitet SE–581 83 Linköping +46 13 28 10 00 , www.liu.se Upphovsrätt Detta dokument hålls tillgängligt på Internet - eller dess framtida ersättare - under 25 år från publicer- ingsdatum under förutsättning att inga extraordinära omständigheter uppstår. Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka ko- pior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervis- ning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säker- heten och tillgängligheten finns lösningar av teknisk och administrativ art. Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsman- nens litterära eller konstnärliga anseende eller egenart. För ytterligare information om Linköping University Electronic Press se förlagets hemsida http://www.ep.liu.se/. Copyright The publishers will keep this document online on the Internet - or its possible replacement - for a period of 25 years starting from the date of publication barring exceptional circumstances. The online availability of the document implies permanent permission for anyone to read, to down- load, or to print out single copies for his/hers own use and to use it unchanged for non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility. According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement. For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its www home page: http://www.ep.liu.se/. Mathilda Moström © Alexander Edberg Students in the 5 year Information Technology program complete a semester-long software develop- ment project during their sixth semester (third year). The project is completed in mid-sized groups, and the students implement a mobile application intended to be used in a multi-actor setting, cur- rently a search and rescue scenario. In parallel they study several topics relevant to the technical and ethical considerations in the project. The project culminates by demonstrating a working product and a written report documenting the results of the practical development process including requirements elicitation. During the final stage of the semester, students create small groups and specialise in one topic, resulting in a bachelor thesis. The current report represents the results obtained during this specialisation work. Hence, the thesis should be viewed as part of a larger body of work required to pass the semester, including the conditions and requirements for a bachelor thesis. Abstract Social networks attract millions of users who want to share information and connect with people. One of those platforms are Twitter,which has the power to greatly shape peo- ple’s opinions and thoughts. It is therefore important to understand how information is shared among users. In this thesis, we characterize the link sharing usage on Twitter, plac- ing particular focus on third-party link shortener services that hide the actual URL from the users until the users click on a generic, shortened URL, focusing mainly on the link man- agement platform Bitly. The purpose of this thesis is to analyze link usage among users over a specific time period, the domains that different users and link shortens direct their users to and compare the click rates of such links with the corresponding retweet rates to see how this vary over time. We use a measurement framework that is developed by two other students from Linköping University to collect datasets over different time periods. First, we will compare a one-week-long dataset from the spring of 2019 to one that is gath- ered one year later in the spring of 2020. Two additional one-week-long datasets has also been collected during the spring of 2020. We use the two main datasets, separated by a year, to evaluate long-term differences, and the three datasets from the spring of 2020 to analyze shorter-term variations in the link usage. The study highlights with this approach is to be able to highlight significant patterns over time, including with regard to what domains that are tweeted. We have found that the usage of URL link shorterns has not decreased over the last year, though the usage of specifically Bitly has done so. The top domains with highest occurrences from 2019 did not get to keep their high rankings in 2020, this is especially true for facebook.com whose occurrence has dropped by 2.7 percentage points in 2020. Our conclusion is that the difference between the years is not huge but that there are some interesting trends and patterns. Given the prevailing pandemic Covid-19, we have also chosen to do a minor analysis of how many users of Twitter link to domains related to this. It turned out that the link sharing of Covid-19 related substances decreased quite sharply during our analysis period. Acknowledgments We would like to thank our supervisor Niklas Carlsson for his support and guidance during the project. We would also like to special thanks Oscar Järpehult and Martin Lindblom for giving us the opportunity to use their framework for our research and for being so helpful answering questions. v Contents Abstract iv Acknowledgments v Contents vi List of Figures viii List of Tables x 1 Introduction 1 1.1 Motivation . 1 1.2 Aim............................................ 1 1.3 Approach . 2 1.4 Contribution . 2 1.5 Delimitations . 2 1.6 Thesis outline . 3 2 Background 4 2.1 Twitter . 4 2.2 Shortened URL . 5 2.3 Top domain ranking sites . 6 2.4 Related work . 6 3 Method 9 3.1 Dataset . 9 3.2 Collection approach . 11 3.3 Limitations . 11 4 Results 12 4.1 High-level link shortener usage . 12 4.2 Domain statistics . 15 4.3 User statistics . 19 4.4 Bitly link interaction . 24 4.5 Verified vs non-verified users . 25 4.6 Covid-19 analysis . 29 5 Discussion 32 5.1 Results . 32 5.2 Method . 33 5.3 The work in a wider context . 33 6 Conclusion 35 6.1 Future work . 36 vi Bibliography 37 A Appendix 40 A.1 URL shorteners . 40 A.2 Collections from 18/3-25/3 and 1/4-8/4 . 43 vii List of Figures 4.1 Top 20 most frequent domains overall (2019). 13 4.2 Top 20 most frequent domains overall (2020). 13 4.3 Top 20 most frequent domains for shortener domains (2019). 14 4.4 Top 20 most frequent domains for shortener domains (2020). 14 4.5 Link popularity distribution to domains of different popularity classes, as defined using the Alexa top-1M lists. 17 4.6 Link popularity distribution to domains of different popularity classes, as defined using the Majestic top-1M lists. 17 4.7 Distribution of domain rank (2019). 18 4.8 Distribution of domain rank (2020). 18 4.9 The results from 2019 is found below the vertical divider in pink and 2020 above in blue. 19 4.10 Distribution of the age for users account at the time of posting their tweet (2019). 20 4.11 Distribution of the age for users account at the time of posting their tweet (2020). 20 4.12 Distribution of the number of tweets favourited by users at the time of posting their tweet (2019). 20 4.13 Distribution of the number of tweets favourited by users at the time of posting their tweet (2020). 21 4.14 Distribution of the number of tweets posted by users at the time of posting their tweet (2019). 21 4.15 Distribution of the number of tweets posted by users at the time of posting their tweet (2020). 21 4.16 Ratio between tweets favourited and tweeted by users at the time of posting their tweet (2019). 22 4.17 Ratio between tweets favourited and tweeted by users at the time of posting their tweet (2020). 22 4.18 Distribution of the number of followers for users at the time of posting their tweet (2019). 22 4.19 Distribution of the number of followers for users at the time of posting their tweet (2020). 23 4.20 Distribution of the number of friends for users at the time of posting their tweet (2019). 23 4.21 Distribution of the number of friends for users at the time of posting their tweet (2020). 23 4.22 Followers-to-friends ratio for users at the time of posting their tweet. 24 4.23 Two scatter plots of Bitly clicks-to-retweets-ratio. 25 4.24 Logarithmic average of Bitly clicks per retweet. 25 4.25 Clicks-to-followers ratio for Bitly links for verified users. 26 4.26 Clicks-to-followers ratio for Bitly links for non-verified users. 26 4.27 Heat-map of retweets vs followers tweeted (2019).