Refine
Document Type
- Master's Thesis (6) (remove)
Language
- English (6) (remove)
Has Fulltext
- yes (6) (remove)
Keywords
- Autosuggestion (1)
- Continuous Evaluation (1)
- Data Mining (1)
- Evaluation Environment (1)
- Kauf (1)
- Lemmatisierung (1)
- LinkedIn (1)
- Longitudinal Evaluation (1)
- Neid (1)
- Politik (1)
As a result of the widespread use of online networking sites, the ways in which people connect and network, both personally and professionally, have been transformed in recent years. Platforms such as LinkedIn or XING have profoundly changed the dynamics of professional networking by providing new means of contact and creating an environment that promotes the exchange of knowledge and ideas. However, compared to social network sites, professional network sites have received little attention in research despite their growing importance. Particularly, the relationship between the use of professional network sites and users' well-being has been understudied. However, the investigation of these platforms is of societal relevance given their consistent growth and the increasing importance of these platforms for both individuals and companies. Existing research on the relationship between the use of social network sites (SNS) and the subjective well-being of users has identified the usage type (active and passive use) as a relevant variable. The aim of this study was to transfer these findings to the context of professional network sites and to explore the relationship between the type of use of a professional network site and the subjective well-being of its users.For this purpose, the active-passive model of SNS use was applied to the context of professional network sites for the first time. To answer the research question, a quantitative online survey was conducted with 526 LinkedIn users. Results of the mediation analyses revealed an indirect positive relation between active use of LinkedIn and well-being. Conversely, a negative indirect relation was found between passive use of LinkedIn and subjective well-being. All tested mediating variables, including social capital for active use and upward social comparison, downward social comparison and envy for passive use, were determined to be relevant in explaining the link between well-being and active and passive LinkedIn use, respectively.
As the information era progresses, the sheer volume of information calls for sophisticated retrieval systems. Evaluating them holds the key to ensuring the reliability and relevance of retrieved information. If evaluated with renowned methods, the measured quality is generally presumed to be dependable. That said, it is often forgotten that most evaluations are only snapshots in time and the reliability might be only valid for a short moment. Further, each evaluation method makes assumptions about the circumstances of a search and thereby has different characteristics. Achieving reliable evaluation is critical to retain the aspired quality of an IR system and maintain the confidence of the users. Therefore, we investigate how the evaluation environment (EE) evolves over time and how this might affect the effectiveness of retrieval systems. Further, attention is paid to the differences in the evaluation methods and how they work together in a continuous evaluation framework. A literature review was conducted to investigate changing components which are then modeled in an extended EE. Exemplarily, the effect of document and qrel updates on the effectiveness of IR systems is investigated through reproducibility experiments in the LongEval shared task. As a result, 11 changing components together with initial measures to quantify how they change are identifed, the temporal consistency of five IR systems could precisely be quantifed through reproducibility and replicability measures and the findings were integrated into a continuous evaluation framework. Ultimately, this work contributes to more holistic evaluations in IR.
Digitalisation is shaping a new consumption era characterised by high connectivity, mobility and a broad range of easily accessible information on products, prices and alternatives. As a result, it becomes more difficult than ever to understand modern consumers along their complex and dynamic path to purchase. However, mobile data about consumers’ behaviour captured on their phone has high potential for facing this challenge. Yet, there is no solution on how to use this data to follow the consumers on their mobile devices. This thesis proposes a first approach on how mobile data collected with smartphone sensing technology can be analysed to assess mobile consumer behaviour along their customer journey. Based on current practices in customer journey analytics, a mobile customer journey model is developed and three analysis concepts are created, which are implemented in an explorative analysis. The results show that mobile sensing data presents a great opportunity for analysing mobile behaviour in three main research areas: examining the touchpoint performance of a brand across mobile apps, describing different target groups by their smartphone usage behaviour and deriving real customer journeys on users’ devices. Nonetheless, further exploration is necessary to unlock the full potential of mobile data in customer journey analytics.
Relevance: Political and private initiatives call for more female founders in start-ups as well as entrepreneurship but with regard to academic research not many studies focused yet on interdisciplinary studies on especially female start-up founders. There is more need to understand the topic to further encourage female founders.
Research question: The research question of this thesis is analysing what kind of patterns can be seen in the entrepreneurial, sociocultural and psychological profile of female founders compared in start-up ecosystems of three different countries, namely Germany, France and Israel?
Approach: I conducted 21 interviews, seven for each city, with a semi-structured guideline focusing on the entrepreneurial, sociocultural and psychological profile. The interviews were transcribed and afterwards analysed by combining the different profiles to find possible patterns. In a final step the observations from each country were compared to one another.
Findings: There are several possible patterns for each country evident. However, a cross-cultural comparison was made difficult by the heterogeneous groups of respondents. It was nevertheless possible to conclude on four crosscultural hypotheses: 1) Female entrepreneurs prefer to work first before starting their own business; 2) The female entrepreneurial profile is risk-taking, purpose-driven, innovative and autonomous; 3) Immigration has a positive effect on the intention to start a business; 4) The majority of female entrepreneurs have a higher education and come from a middle to upper social class.
Research data which is put into long term storage needs to have quality metadata attached so it may be found in the future. Metadata facilitates the reuse of data by third parties and makes it citable in new research contexts and for new research questions. However, better tools are needed to help the researchers add metadata and prepare their data for publication. These tools should integrate well in the existing research workflow of the scientists, to allow metadata enrichment even while they are creating, gathering or collecting the data. In this thesis an existing data publication tool from the project DARIAH-DE was connected to a proven file synchronization software to allow the researchers prepare the data from their personal computers and mobile devices and make it ready for publication. The goal of this thesis was to find out whether the use of file synchronization software eases the data publication process for the researchers.
Analysing the systematics of search engine autocompletion functions by means of data mining methods
(2017)
In the internet era, the information that can be found about politicians online can influence
events such as the results of elections. Research has shown that biased search rankings can
shift the voting preferences of undecided voters. This shows the importance of studying
online search behaviour, especially in the pre-elections phase, when search results can
have a particular influence on the future political scene of a country.
This master thesis aimed to study the behaviour of online search engines in a period before
the German federal election in 2017. The aim was to ascertain if there is any pattern to be
found in the auto-suggestions for searches related to politicians.
In order to gather data for this experiment, a crawler browsed search engine web pages,
input a name and a surname of a politician, and saved that together with all autosuggestions
from the search engine. The autosuggestions were prepared for the analysis and
divided into semantic groups with the help of clustering algorithms.
Different statistical methods, such as correlation analysis, regression analysis, and clustering
were used to identify patterns in the data. The research showed that there are
no particularly strong patterns in the autosuggestions for searches related to politician’s
names. Only moderate dependence was found between gender and personal topics, and
showed that a higher amount of personal information autosuggestions correspond more
to female politicians.