Ph.D. Theses
Permanent URI for this collection
Browse
Recent Submissions
Item Developing a dynamic predictive policing system(Thesis (Ph.D.) - Bogazici University. Institute for Graduate Studies in the Social Sciences, 2022., 2022.) Hakyemez, Tuğrul Cabir.; Badur, Bertan Yılmaz.The retrospective predictive policing techniques are atheoretical and therefore remain incapable of sensing the changing crime risk across the streets. In this study, we aim to develop a dynamic predictive policing system that capitalizes on theory-based risk indicators. The sample includes all the theft and robbery incidents in Chicago between 2014-2019. In the first step, pipelining bivariate network K analysis and segmented regression, we introduce novel distance-aware risk functions that operationalize spatiotemporal crime risk around the selected urban features (i.e., bus stop, fast food restaurant, gas station, grocery store, pub). In the second step, we develop various network-based predictive policing methods using graph-based deep learning algorithms (i.e., GraphWavenet, Spatiotemporal Graph Convolutional Networks). These methods generate weekly and intraday hotspot predictions. We complement these methods with various theory-based risk indicators including a risk score devised from the novel risk functions, 311 calls, park events, and cooccurring crime incidents. The results showcase that crime risk around urban features varies across space, time, and crime types. Furthermore, this risk is found to be significantly correlated with the regional socioeconomic characteristics. Another important result shows that incorporating theory based indicators improved the performance of the retrospective methods up to 68%. Amongst the algorithms, GraphWavenet is found to outperform its counterparts in the majority of the prediction models with an accuracy as high as 80%. The proposed system helps law enforcement agents in planning their operations efficiently by pinpointing the micro geographical units with relatively higher risks in the next time step.Item A digital innovations-driven regeneration model and corporate sustainability(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in the Social Sciences, 2020., 2020.) Coşkun Setirek, Abide.; Tanrıkulu, Zuhal.The traditional ways of doing business have been changed by digital innovations such as the Internet of things, blockchain and digital currency, data analytics, artificial intelligence, robots, additive manufacturing, etc. Firms can stay competitive using the benefits of digital technologies. The spread of the coronavirus disease in 2019 (COVID-19) all over the world has created a better understanding of the importance of organizations’ ability to keep up with digital innovations. In this study, a method for digital innovations-driven business model regeneration is developed and a dynamic business model, which can also be used in the business model regeneration process, to examine the effects of digital innovation strategies on the corporate sustainability is proposed. For this purpose, the existing literature on the business model innovation and system dynamic are examined, and the empirical data are collected from 44 managers using semi-structured interviews to complement gaps in the literature. Moreover, the digital innovations-driven business model regeneration method, which is proposed in this study, is applied to a real case. This study extends the literature on the business model innovation and the dynamic business model. The study can provide strategy analysts and managers with an opportunity to analyze the effects of potential digital innovation strategies on their current business models and to explore the most effective digital innovation strategies in order to regenerate their business model to gain a competitive advantage over their competitors or to sustain their business in light of technological developments.Item Essays in learning representations of complex networks(Thesis (Ph.D.) - Bogazici University. Institute for Graduate Studies in the Social Sciences, 2021., 2021.) Gürsoy, Furkan.; Badur, Bertan Yılmaz.This thesis contains three essays in learning representation of complex networks, the first two of which develop new methods and the third utilizes these methods in a real-world application. The first essay provides methods for extracting underlying signed network backbones from intrinsically dense weighted networks. Utilizing a null model based on statistical techniques, we propose significance and vigor filters that enable inferring edge signs and weights. Empirical analysis on four real-world networks reveals that the proposed filters extract meaningful and sparse signed backbones that exhibit characteristics typically associated with signed networks while respecting the multiscale nature of the network. The second essay deals with the misalignment problem in dynamic representation learning. We provide the first formal definitions of alignment and stability, propose novel metrics for measuring them, and show their suitability through a set of synthetic and real-world experiments. We show that, by ensuring alignment, the performance of dynamic network inference tasks improves by a remarkable amount. The third essay applies the novel methods developed in the first two essays as well as other methods from the network analysis literature to investigate the structure and dynamics of internal migration in Turkey. In addition to providing unique and specific insights, we find that most migration links are geographically bounded with exceptions of cities with large economic activity, migration takes place in well-defined routes, counter-streams develop for major migration streams, and the migration system is largely stable over time; which are generally in line with classical migration laws.Item A social media big data mining framework for detecting sentiments in multiple languages(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in the Social Sciences, 2018., 2018.) Coşkun, Mustafa.; Özturan, Meltem.The popularity of social media platforms has generated a new social interaction environment thus a new collaboration network among individuals. These platforms own tremendous amount of data about users’ behaviors and sentiments. One of these platforms is Twitter, which provides researchers data potential of benefit for their studies. Based on Twitter data, in this study a multilingual sentiment detection framework is proposed to compute European Gross National Happiness (GNH). This framework consists of a novel data collection, filtering and sampling method, and multilingual sentiment detection algorithm for social media big data, and tested with nine European countries (United Kingdom, Germany, Sweden, Turkey, Portugal, Netherlands, Italy, France and Spain) and their national languages over six-year period. The reliability of the data is checked with peak/troughs comparison for special days from Wikipedia. The validity is checked with a group of correlation analyses with OECD Life Satisfaction survey reports’, currency exchanges, and national stock market time series data. Then, the European GNH map is drawn for six years. Lastly, an exploratory study for determining the relationships between users’ Twitter account features (number of tweets, number of followers etc.) and happiness polarities are analyzed. Main aim of this study is to propose a novel multilingual social media sentiment analysis framework for calculating GNH for countries and change the way of OECD type organizations’ survey and interview methodology. Also, it is believed that this framework can serve more detailed results (e.g. daily or hourly sentiments of society in different languages).Item Developing a context-aware location recommender system for location-based social networks(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in the Social Sciences, 2018., 2018.) Bozanta, Aysun.; Kutlu, Birgül.People think about where to go many times throughout their lives. Although it is a very rapid and repetitive decision, generally it is hard to choose suitable places from endless number of options for some specific circumstances. Recommender systems are supposed to help to deal with those issues and take appropriate actions. However, the location decision is different from other decisions like what to listen, buy, or read from various aspects. The popularity of location-based social networks has prompted researchers to study recommendation systems for location. Traditional recommendation algorithms have been used for location recommendation. When used separately, each venue recommendation system algorithm has drawbacks. Another issue is that the context information is not commonly used in venue recommendation systems. Time, distance and weather conditions have more impact on decisions about where to go than all other decisions. Another point that should not be disregarded is that the effects of those contextual variables differ from user to user. This study proposes a hybrid recommendation model that combines contextual information, user- and item-based collaborative filtering and content-based filtering. For this purpose, user visit histories, venue-related information and contextual information related to individual user visits were collected from Twitter, Foursquare, and Weather Underground. The proposed hybrid system is evaluated using both offline experiments and a user study. This proposed system shows better results than baseline approaches.Item The anatomy of an online social community network(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in the Social Sciences, 2018., 2018.) Akar, Ezgi.; Mardikyan, Sona Kunuzyan.The emergence of Web 2.0 has revolutionized the ways of communication on the Internet and has allowed people to form their virtual worlds involving online communities and social networks. People have started to generate their contents, share them, and communicate with each other in these communities and networks. In parallel to the generation of huge amount of contents in these platforms, these communities and networks have become valuable data sources for businesses. It is the fact that analysis of user content in online communities and social networks allows businesses to enhance their business value and achieve their goals. In this sense, to create and manage these communities successfully, managers need to understand how to motivate community members and keep them frequently involved. Therefore, this study employs social network analysis to map and understand the network structure of an online community and to detect sub-communities in it. Additionally, it explores and identifies user roles in an online community and proposes a research model that investigates members’ usage intentions of the community. The research model also analyzes the moderating effect of these investigated user roles on members’ usage intentions of the community. In this manner, this study combines various theories for a better understanding of what roles exist in online communities, what roles members prefer to adopt, what usage intentions members have by presenting a four-phase methodology. Additional to theoretical implications, the study also guides managers to develop motivational strategies to keep their members continually satisfied in online communities.