Matching Items (113)

Filtering by

Clear all filters

149714-Thumbnail Image.png

Analyzing the dynamics of communication in online social networks

Description

This thesis deals with the analysis of interpersonal communication dynamics in online social networks and social media. Our central hypothesis is that communication dynamics between individuals manifest themselves via three key aspects: the information that is the content of communication,

This thesis deals with the analysis of interpersonal communication dynamics in online social networks and social media. Our central hypothesis is that communication dynamics between individuals manifest themselves via three key aspects: the information that is the content of communication, the social engagement i.e. the sociological framework emergent of the communication process, and the channel i.e. the media via which communication takes place. Communication dynamics have been of interest to researchers from multi-faceted domains over the past several decades. However, today we are faced with several modern capabilities encompassing a host of social media websites. These sites feature variegated interactional affordances, ranging from blogging, micro-blogging, sharing media elements as well as a rich set of social actions such as tagging, voting, commenting and so on. Consequently, these communication tools have begun to redefine the ways in which we exchange information, our modes of social engagement, and mechanisms of how the media characteristics impact our interactional behavior. The outcomes of this research are manifold. We present our contributions in three parts, corresponding to the three key organizing ideas. First, we have observed that user context is key to characterizing communication between a pair of individuals. However interestingly, the probability of future communication seems to be more sensitive to the context compared to the delay, which appears to be rather habitual. Further, we observe that diffusion of social actions in a network can be indicative of future information cascades; that might be attributed to social influence or homophily depending on the nature of the social action. Second, we have observed that different modes of social engagement lead to evolution of groups that have considerable predictive capability in characterizing external-world temporal occurrences, such as stock market dynamics as well as collective political sentiments. Finally, characterization of communication on rich media sites have shown that conversations that are deemed "interesting" appear to have consequential impact on the properties of the social network they are associated with: in terms of degree of participation of the individuals in future conversations, thematic diffusion as well as emergent cohesiveness in activity among the concerned participants in the network. Based on all these outcomes, we believe that this research can make significant contribution into a better understanding of how we communicate online and how it is redefining our collective sociological behavior.

Contributors

Agent

Created

Date Created
2011

153872-Thumbnail Image.png

Making thin data thick: user behavior analysis with minimum information

Description

With the rise of social media, user-generated content has become available at an unprecedented scale. On Twitter, 1 billion tweets are posted every 5 days and on Facebook, 20 million links are shared every 20 minutes. These massive collections of

With the rise of social media, user-generated content has become available at an unprecedented scale. On Twitter, 1 billion tweets are posted every 5 days and on Facebook, 20 million links are shared every 20 minutes. These massive collections of user-generated content have introduced the human behavior's big-data.

This big data has brought about countless opportunities for analyzing human behavior at scale. However, is this data enough? Unfortunately, the data available at the individual-level is limited for most users. This limited individual-level data is often referred to as thin data. Hence, researchers face a big-data paradox, where this big-data is a large collection of mostly limited individual-level information. Researchers are often constrained to derive meaningful insights regarding online user behavior with this limited information. Simply put, they have to make thin data thick.

In this dissertation, how human behavior's thin data can be made thick is investigated. The chief objective of this dissertation is to demonstrate how traces of human behavior can be efficiently gleaned from the, often limited, individual-level information; hence, introducing an all-inclusive user behavior analysis methodology that considers social media users with different levels of information availability. To that end, the absolute minimum information in terms of both link or content data that is available for any social media user is determined. Utilizing only minimum information in different applications on social media such as prediction or recommendation tasks allows for solutions that are (1) generalizable to all social media users and that are (2) easy to implement. However, are applications that employ only minimum information as effective or comparable to applications that use more information?

In this dissertation, it is shown that common research challenges such as detecting malicious users or friend recommendation (i.e., link prediction) can be effectively performed using only minimum information. More importantly, it is demonstrated that unique user identification can be achieved using minimum information. Theoretical boundaries of unique user identification are obtained by introducing social signatures. Social signatures allow for user identification in any large-scale network on social media. The results on single-site user identification are generalized to multiple sites and it is shown how the same user can be uniquely identified across multiple sites using only minimum link or content information.

The findings in this dissertation allows finding the same user across multiple sites, which in turn has multiple implications. In particular, by identifying the same users across sites, (1) patterns that users exhibit across sites are identified, (2) how user behavior varies across sites is determined, and (3) activities that are observed only across sites are identified and studied.

Contributors

Agent

Created

Date Created
2015

153091-Thumbnail Image.png

Planning challenges in human-robot teaming

Description

As robotic technology and its various uses grow steadily more complex and ubiquitous, humans are coming into increasing contact with robotic agents. A large portion of such contact is cooperative interaction, where both humans and robots are required to work

As robotic technology and its various uses grow steadily more complex and ubiquitous, humans are coming into increasing contact with robotic agents. A large portion of such contact is cooperative interaction, where both humans and robots are required to work on the same application towards achieving common goals. These application scenarios are characterized by a need to leverage the strengths of each agent as part of a unified team to reach those common goals. To ensure that the robotic agent is truly a contributing team-member, it must exhibit some degree of autonomy in achieving goals that have been delegated to it. Indeed, a significant portion of the utility of such human-robot teams derives from the delegation of goals to the robot, and autonomy on the part of the robot in achieving those goals. In order to be considered truly autonomous, the robot must be able to make its own plans to achieve the goals assigned to it, with only minimal direction and assistance from the human.

Automated planning provides the solution to this problem -- indeed, one of the main motivations that underpinned the beginnings of the field of automated planning was to provide planning support for Shakey the robot with the STRIPS system. For long, however, automated planners suffered from scalability issues that precluded their application to real world, real time robotic systems. Recent decades have seen a gradual abeyance of those issues, and fast planning systems are now the norm rather than the exception. However, some of these advances in speedup and scalability have been achieved by ignoring or abstracting out challenges that real world integrated robotic systems must confront.

In this work, the problem of planning for human-hobot teaming is introduced. The central idea -- the use of automated planning systems as mediators in such human-robot teaming scenarios -- and the main challenges inspired from real world scenarios that must be addressed in order to make such planning seamless are presented: (i) Goals which can be specified or changed at execution time, after the planning process has completed; (ii) Worlds and scenarios where the state changes dynamically while a previous plan is executing; (iii) Models that are incomplete and can be changed during execution; and (iv) Information about the human agent's plan and intentions that can be used for coordination. These challenges are compounded by the fact that the human-robot team must execute in an open world, rife with dynamic events and other agents; and in a manner that encourages the exchange of information between the human and the robot. As an answer to these challenges, implemented solutions and a fielded prototype that combines all of those solutions into one planning system are discussed. Results from running this prototype in real world scenarios are presented, and extensions to some of the solutions are offered as appropriate.

Contributors

Agent

Created

Date Created
2014

153926-Thumbnail Image.png

Leveraging collective wisdom in a multilabeled blog categorization environment

Description

One of the most remarkable outcomes resulting from the evolution of the web into Web 2.0, has been the propelling of blogging into a widely adopted and globally accepted phenomenon. While the unprecedented growth of the Blogosphere has added diversity

One of the most remarkable outcomes resulting from the evolution of the web into Web 2.0, has been the propelling of blogging into a widely adopted and globally accepted phenomenon. While the unprecedented growth of the Blogosphere has added diversity and enriched the media, it has also added complexity. To cope with the relentless expansion, many enthusiastic bloggers have embarked on voluntarily writing, tagging, labeling, and cataloguing their posts in hopes of reaching the widest possible audience. Unbeknown to them, this reaching-for-others process triggers the generation of a new kind of collective wisdom, a result of shared collaboration, and the exchange of ideas, purpose, and objectives, through the formation of associations, links, and relations. Mastering an understanding of the Blogosphere can greatly help facilitate the needs of the ever growing number of these users, as well as producers, service providers, and advertisers into facilitation of the categorization and navigation of this vast environment. This work explores a novel method to leverage the collective wisdom from the infused label space for blog search and discovery. The work demonstrates that the wisdom space can provide a most unique and desirable framework to which to discover the highly sought after background information that could aid in the building of classifiers. This work incorporates this insight into the construction of a better clustering of blogs which boosts the performance of classifiers for identifying more relevant labels for blogs, and offers a mechanism that can be incorporated into replacing spurious labels and mislabels in a multi-labeled space.

Contributors

Agent

Created

Date Created
2015

154392-Thumbnail Image.png

A Case Study of Executive Stock Compensation Design for The State-Owned Firms

Description

Executive compensation design involving equity shares has been widely used in Europe, the United States and other developed countries where the capital markets are relatively mature. In China, due to the differences in industries, ownership structure, stages of enterprise development,

Executive compensation design involving equity shares has been widely used in Europe, the United States and other developed countries where the capital markets are relatively mature. In China, due to the differences in industries, ownership structure, stages of enterprise development, constraints faced by the firms, the executive compensation design using equity shares tends to vary accordingly. For the state-owned companies, the situations are more complex than others. This complexity has not been a focus of the past literature, particularly on the compensation contract design and its subsequent implementation. Based on Coase contract theorem, agency theory and human capital theory, I examined how different state-owned firms vary in their approaches on managerial stock compensation design using a case study approach. The thesis concludes with a summary of major findings and a discussion of policy implications.

Contributors

Agent

Created

Date Created
2016

154403-Thumbnail Image.png

Visual analytics for spatiotemporal cluster analysis

Description

Traditionally, visualization is one of the most important and commonly used methods of generating insight into large scale data. Particularly for spatiotemporal data, the translation of such data into a visual form allows users to quickly see patterns, explore summaries

Traditionally, visualization is one of the most important and commonly used methods of generating insight into large scale data. Particularly for spatiotemporal data, the translation of such data into a visual form allows users to quickly see patterns, explore summaries and relate domain knowledge about underlying geographical phenomena that would not be apparent in tabular form. However, several critical challenges arise when visualizing and exploring these large spatiotemporal datasets. While, the underlying geographical component of the data lends itself well to univariate visualization in the form of traditional cartographic representations (e.g., choropleth, isopleth, dasymetric maps), as the data becomes multivariate, cartographic representations become more complex. To simplify the visual representations, analytical methods such as clustering and feature extraction are often applied as part of the classification phase. The automatic classification can then be rendered onto a map; however, one common issue in data classification is that items near a classification boundary are often mislabeled.

This thesis explores methods to augment the automated spatial classification by utilizing interactive machine learning as part of the cluster creation step. First, this thesis explores the design space for spatiotemporal analysis through the development of a comprehensive data wrangling and exploratory data analysis platform. Second, this system is augmented with a novel method for evaluating the visual impact of edge cases for multivariate geographic projections. Finally, system features and functionality are demonstrated through a series of case studies, with key features including similarity analysis, multivariate clustering, and novel visual support for cluster comparison.

Contributors

Agent

Created

Date Created
2016

154410-Thumbnail Image.png

Investment Style and Performance Attribution Analysis on Chinese A Share Market

Description

With the fast development of Chinese capital market, an increasing number of institutions and retail investors invest through professional managers. The key to evaluating investment manager’s skill and performance persistence largely lies in portfolio style research and attribution analysis.

With the fast development of Chinese capital market, an increasing number of institutions and retail investors invest through professional managers. The key to evaluating investment manager’s skill and performance persistence largely lies in portfolio style research and attribution analysis.

The current dissertation takes advantage of a unique dataset, uncover hidden investment style and trading behavior, understanding their source of excess returns, and establishing a more comprehensive methodology for evaluating portfolio performance and manager skills.

The dissertation focuses on quantitative analysis. Highlights three most important aspects. Investment style determines the systematic returns and risks of any portfolio, and can be assessed ex-ante; Transaction can be observed and modified during the investment process; and return attribution can be implemented to evaluate portfolio (managers), ex-post. Hence, these three elements make up a comprehensive and logical investment process.

Investment style is probably the most important factor in determining portfolio returns. However, Chinese investment managers are under constant pressure to follow the market trend and shift style accordingly. Therefore, accurately identifying and predicting each manager’s investment style proves critically valuable.

In addition, transaction data probably provides the most reliable source of information in observing and evaluating an investment manager’s style and strategy, in the middle of the investment process.

Despite the efficacy of traditional return attribution methodology, there are clear limitations. The current study proposes a novel return attribution methodology, by synthesizing major portfolio strategy components, such as risk exposure adjustment, sector rotation, stock selection, altogether. Our novel methodology reveals that investment managers do not obtain much abnormal returns through risk exposure adjustment or sector rotation. Instead, Chinese investment managers seem to enjoy most of their excess returns through stock selection.

In addition, we find several interesting patterns in Chinese A-share market: 1). There is a negative relationship between asset under management (AUM) and investment performance, beyond certain AUM threshold; 2). There are limited benefits from style switching in the long run; 3). Many investment managers use CSI 300 component stocks as portfolio ballast and speculate with CSI500 and Medium-and-Small board component stocks for excess returns; 4). There is no systematic negative relationship between portfolio turnover and investment performance; despite negative relationship within certain sub-samples and sectors; 5). It is plausible to construct out-performing portfolios with style index funds and ETFs.

Contributors

Agent

Created

Date Created
2016

154374-Thumbnail Image.png

A Case Study of Credit Risk Analysis and Modeling for SMEs -In an Internet Finance Setting

Description

In the last two years, China’s booming of Internet Finance Platform made significant impacts on three dimensions. Compared with the conventional market, Internet Finance is asserted to open a revolutionary pathway of lending where by small and mid-sized companies may

In the last two years, China’s booming of Internet Finance Platform made significant impacts on three dimensions. Compared with the conventional market, Internet Finance is asserted to open a revolutionary pathway of lending where by small and mid-sized companies may overcome the financing dilemma on credit accessibility and high cost. In other words, Internet Finance is hyped to be able to reduce information asymmetry, enhance allocation efficiency of resources, and promote product and process innovations for the financial institutions. However, the core essence of Internet Finance rests on risk assessment and control – a fundamental element applies to all forms of financing. Most current practice of internet finance on risk assessment and control remains unchanged from the mindset of traditional banking practices for small and medium sized firms. Hence, the same problems persisted and may only become even worse under the internet finance platform if no innovations take place.

In this thesis, the author proposed and tested a credit risk assessment model using data analytics techniques through an in-depth cases study with actual transaction data. Specifically, based on the 30,000 observations collected from actual transactional data from small and medium size firms of China’s home furnishing industry. The preliminary results are promising in spite of the limitations. The thesis concludes with the findings of relevance to improve the current practices and suggests areas of future research.

Contributors

Agent

Created

Date Created
2016

154377-Thumbnail Image.png

Research on Factors Influencing Individual’s Behavior of Energy Management

Description

With the rapid rise of distributed generation, Internet of Things, and mobile Internet, both U.S. and European smart home manufacturers have developed energy management solutions for individual usage. These applications help people manage their energy consumption more efficiently. Domestic manufacturers

With the rapid rise of distributed generation, Internet of Things, and mobile Internet, both U.S. and European smart home manufacturers have developed energy management solutions for individual usage. These applications help people manage their energy consumption more efficiently. Domestic manufacturers have also launched similar products.

This paper focuses on the factors influencing Energy Management Behaviour (EMB) at the individual level. By reviewing academic literature, conducting surveys in Beijing, Shanghai and Guangzhou, the author builds an integrated behavioural energy management model of the Chinese energy consumers. This paper takes the vague term of EMB and redefines it as a function of two separate behavioural concepts: Energy Management Intention (EMI), and the traditional Energy Saving Intention (ESI).

Secondly, the author conducts statistical analyses on these two behavioural concepts. EMI is the main driver behind an individual’s EMB. EMI is affected by Behavioural Attitudes, Subjective Norms, and Perceived Behavioural Control (PBC). Among these three key factors, PBC exerts the strongest influence. This implies that the promotion of the energy management concept is mainly driven by good application user experience (UX). The traditional ESI also demonstrates positive influence on EMB, but its impact is weaker than the impacts arising under EMI’s three factors. In other words, the government and manufacturers may not be able to change an individual's energy management behaviour if they rely solely on their traditional promotion strategies. In addition, the study finds that the government may achieve better promotional results by launching subsidies to the manufacturers of these kinds of applications and smart appliances.

Contributors

Agent

Created

Date Created
2016

151129-Thumbnail Image.png

Trust and profit sensitive ranking for the deep web and on-line advertisements

Description

Ranking is of definitive importance to both usability and profitability of web information systems. While ranking of results is crucial for the accessibility of information to the user, the ranking of online ads increases the profitability of the search provider.

Ranking is of definitive importance to both usability and profitability of web information systems. While ranking of results is crucial for the accessibility of information to the user, the ranking of online ads increases the profitability of the search provider. The scope of my thesis includes both search and ad ranking. I consider the emerging problem of ranking the deep web data considering trustworthiness and relevance. I address the end-to-end deep web ranking by focusing on: (i) ranking and selection of the deep web databases (ii) topic sensitive ranking of the sources (iii) ranking the result tuples from the selected databases. Especially, assessing the trustworthiness and relevances of results for ranking is hard since the currently used link analysis is inapplicable (since deep web records do not have links). I formulated a method---namely SourceRank---to assess the trustworthiness and relevance of the sources based on the inter-source agreement. Secondly, I extend the SourceRank to consider the topic of the agreeing sources in multi-topic environments. Further, I formulate a ranking sensitive to trustworthiness and relevance for the individual results returned by the selected sources. For ad ranking, I formulate a generalized ranking function---namely Click Efficiency (CE)---based on a realistic user click model of ads and documents. The CE ranking considers hitherto ignored parameters of perceived relevance and user dissatisfaction. CE ranking guaranteeing optimal utilities for the click model. Interestingly, I show that the existing ad and document ranking functions are reduced forms of the CE ranking under restrictive assumptions. Subsequently, I extend the CE ranking to include a pricing mechanism, designing a complete auction mechanism. My analysis proves several desirable properties including revenue dominance over popular Vickery-Clarke-Groves (VCG) auctions for the same bid vector and existence of a Nash equilibrium in pure strategies. The equilibrium is socially optimal, and revenue equivalent to the truthful VCG equilibrium. Further, I relax the independence assumption in CE ranking and analyze the diversity ranking problem. I show that optimal diversity ranking is NP-Hard in general, and that a constant time approximation algorithm is not likely.

Contributors

Agent

Created

Date Created
2012