Projects
Where I am coming from
I studied computer science at the University of Potsdam in Germany. Already during the last years at the university, I was interested in analyzing online behavior, whereby I focused on user interactions in collaborative online applications such as wikis and folksonomies. As part of a team that developed a software for the analysis of such applications (SONIVIS), I was responsible for the integration of a text mining component. This component enabled us to study the collaborative creation of content in such applications.
Afterwards, I was working as a research scientist at the ISI Foundation in Turin where I could apply my acquired skills and knowledge while analyzing users' attention patterns in Twitter.
Where I am now
With the beginning of my Ph.D. in 2011 and the position at Yahoo linked with it, it was possible for me to study online behavior in a broader context, considering a much larger set of web application of various types. In addition, my interests moved primarily to analyzing users' browsing behavior on websites to characterize how users engage with them. In the online world, user engagement refers to the quality of the user experience that emphasizes the positive aspects of the interaction with a website and, in particular, the phenomena associated with wanting to use that website longer and frequently. Nowadays, successful websites are not just used, they are engaged with.
However, before we can design engaging web applications, it is crucial that we are able to measure user engagement. A widely-used approach to measure engagement is through web analytics aiming at assessing users' depth of interaction with a website. Web analytics include the usage of online behavior metrics such as click-through rate, time spent on a site (dwell time), page views, return rates, and number of users. At the beginning of my Ph.D., I studied existing online behavior metrics, focusing particularly on their limitations and proposed new metrics that expose so far unconsidered aspects of user engagement. My position at Yahoo gave me the possibility to collaborate with highly qualified people and to develop and evaluate these metrics using large collections of data. We carried out two projects resulting in the following sets of user engagement metrics:
Where I want to go
I am deeply interested in applying my acquired skills and knowledge on further web applications with the objective to find solutions to increase engagement and revenue. More precisely, I want to explore the browsing behavior of users on the corresponding website, and, if necessary, develop methods and metrics that are tailored for it. I then want to use this to identify well and poor performing features and pages, to explore possible navigation difficulties, to compare casual with engaged users, and so on, and finally make suggestions on how all of this can be used to optimize the website.
I also want to go one step further and apply my findings. This means that I want to perform changes on the website to improve the quality of the website, and evaluate the impact of such changes using web analytics in combination with A/B testing, questionnaires, usability tests, etc.
I studied computer science at the University of Potsdam in Germany. Already during the last years at the university, I was interested in analyzing online behavior, whereby I focused on user interactions in collaborative online applications such as wikis and folksonomies. As part of a team that developed a software for the analysis of such applications (SONIVIS), I was responsible for the integration of a text mining component. This component enabled us to study the collaborative creation of content in such applications.
Afterwards, I was working as a research scientist at the ISI Foundation in Turin where I could apply my acquired skills and knowledge while analyzing users' attention patterns in Twitter.
Where I am now
With the beginning of my Ph.D. in 2011 and the position at Yahoo linked with it, it was possible for me to study online behavior in a broader context, considering a much larger set of web application of various types. In addition, my interests moved primarily to analyzing users' browsing behavior on websites to characterize how users engage with them. In the online world, user engagement refers to the quality of the user experience that emphasizes the positive aspects of the interaction with a website and, in particular, the phenomena associated with wanting to use that website longer and frequently. Nowadays, successful websites are not just used, they are engaged with.
However, before we can design engaging web applications, it is crucial that we are able to measure user engagement. A widely-used approach to measure engagement is through web analytics aiming at assessing users' depth of interaction with a website. Web analytics include the usage of online behavior metrics such as click-through rate, time spent on a site (dwell time), page views, return rates, and number of users. At the beginning of my Ph.D., I studied existing online behavior metrics, focusing particularly on their limitations and proposed new metrics that expose so far unconsidered aspects of user engagement. My position at Yahoo gave me the possibility to collaborate with highly qualified people and to develop and evaluate these metrics using large collections of data. We carried out two projects resulting in the following sets of user engagement metrics:
- Online multitasking:
Users often access and re-access more than one site during an online session,
effectively engaging in online multitasking. We studied the effect of multitasking
on engagement, and we defined new metrics that characterize such behavior.
J. Lehmann, M. Lalmas, G. Dupret, and R. Baeza-Yates.
[cikm 2013|slideshare|labtomarket blog|poster] - Inter-site engagement:
Many large online providers (e.g., Amazon, Google, Yahoo) offer a variety of websites, ranging from
shopping to news. Standard engagement metrics are not able to assess engagement with more than one
website. We proposed an approach for measuring engagement with a network of sites by accounting for
the traffic between the sites.
J. Lehmann, M. Lalmas, and R. Baeza-Yates.
[book: big data analytics 2015|cikm workshop 2013|slideshare]
- News sites: We study story-focused news reading, which occurs when users read several articles related to a particular news development. Our aim is to understand the effect of story-focused reading on engagement and how news sites can support this phenomenon. J. Lehmann, C. Castillo, M. Lalmas, R. Baeza-Yates. - Ongoing project
- Wikidata: In this project, we analyze users editing activities in Wikidata to discover participation patterns that the community exhibit. We hope that our study informs future analyses and developments and, as a result, allows us to build better tools to support contributors in peer-production-based ontology engineering. C. Müller-Birn, B. Karran, J. Lehmann, and M. Luczak-Rösch. [opensym 2015] - Ongoing project
- Native advertising: The aim of this project was to understand how users experience adverts and how the ad quality impact user engagement with the publisher site. We also developed a prediction model (now implemented in Yahoo Gemini) to identify high quality ads by analyzing their landing pages and relating these to the ad post-click experience. M. Lalmas, J. Lehmann, G. Shaked, F. Silvestri and G. Tolomei. [kdd 2015]
- Wikipedia: We studied how readers engage with Wikipedia by characterizing their reading preferences and behaviors, and illustrated how reader engagement can provide valuable insights to Wikipedia's editor community. J. Lehmann, C. Müller-Birn, D. Laniado, M. Lalmas, and A. Kaltenbrunner. [ht 2014|wikimania 2014|slideshare]
- Twitter: We developed an application that can help news providers, their journalists and editors, in discovering story-related content and their curators in Twitter. J. Lehmann, C. Castillo, M. Lalmas. [icwsm 2013|www companion 2013|slideshare|slides]
Where I want to go
I am deeply interested in applying my acquired skills and knowledge on further web applications with the objective to find solutions to increase engagement and revenue. More precisely, I want to explore the browsing behavior of users on the corresponding website, and, if necessary, develop methods and metrics that are tailored for it. I then want to use this to identify well and poor performing features and pages, to explore possible navigation difficulties, to compare casual with engaged users, and so on, and finally make suggestions on how all of this can be used to optimize the website.
I also want to go one step further and apply my findings. This means that I want to perform changes on the website to improve the quality of the website, and evaluate the impact of such changes using web analytics in combination with A/B testing, questionnaires, usability tests, etc.
Links
News
@ 2012 Janette Lehmann.
Disclaimer: The author of this web page is not responsible for the contents of webpages linked through this web page
and created by other authors. The views and opinions expressed in the linked web pages authored by others are those of the authors.
Disclaimer: The author of this web page is not responsible for the contents of webpages linked through this web page
and created by other authors. The views and opinions expressed in the linked web pages authored by others are those of the authors.