Showing papers by "Peter A. Gloor published in 2009"

Capturing Individual and Group Behavior with Wearable Sensors

Abstract: We show how to obtain high level descriptions of human behavior in terms of physical activity, speech activity, face-to-face interaction (f2f), physical proximity, and social network attributes from sensor data. We present experimental results that showthat it is possible to identify individual personality traits as well as subjective and objective group performance metrics from low level data collected using wearable sensors.

Web Science 2.0: Identifying Trends through Semantic Social Network Analysis

Abstract: We introduce a novel set of social network analysis based algorithms for mining the Web, blogs, and online forums to identify trends and find the people launching these new trends. These algorithms have been implemented in Condor, a software system for predictive search and analysis of the Web and especially social networks. Algorithms include the temporal computation of network centrality measures, the visualization of social networks as Cybermaps, a semantic process of mining and analyzing large amounts of text based on social network analysis, and sentiment analysis and information filtering methods. The temporal calculation of betweenness of concepts permits to extract and predict long-term trends on the popularity of relevant concepts such as brands, movies, and politicians. We illustrate our approach by qualitatively comparing Web buzz and our Web betweenness for the 2008 US presidential elections, as well as correlating the Web buzz index with share prices.

Wearable sensors for pervasive healthcare management

Abstract: We show that it is possible to identify individual personality traits and measure group performance in a Post-anesthesia Care Unit (PACU) using wearable sensors.We instrumented a group of 67 nurses working in the PACU of a Boston area hospital with sociometric badges capable of measuring physical activity, speech activity, face-to-face interaction, and physical proximity. Using the data collected with these sensors we were able to estimate the daily average length of stay (LOS) and number of delays.

Analysis of Informal Communication Networks – A Case Study

Abstract: The structure and dynamics of informal communication networks are of central importance for the functionality of enterprise workflows and for performance and innovation of knowledge-centric organizations. While most executives are aware of this fact, there is a general lack of (semi-) automated, IT-supported methods and instruments to make informal communication networks measurable. Although logging of electronic communications has made considerable progress over the past few years, it is still extremely difficult to map personal interaction; manual approaches in particular are extremely error-prone. The article shows how informal communication networks can be investigated by IT-based methods. At the same time, the authors will be presenting an instrument (“Social Badges”) that collects personal communications automatically and more precisely than legacy approaches allow. The practical applicability of the approach is evaluated through a case study.

The Virtual Mirror - Reflecting on Social and Psychological Self to Increase Organizational Creativity

Abstract: This article offers a novel approach for managers to increase creativity in their teams by measuring individual personality characteristics, crucial for creative people. We assess inter-personal interaction with sensor-equipped badges worn on the body. In a research project with 22 study subjects, who wore the badges during work for one month, we were able to predict extroversion, neuroticism, openness, and agreeability based on microscopic social network analysis. We obtained control measures of these values with a standard psychological test (NEO-FFI). As opposed to conventional personality tests, where people have to fill out lengthy questionnaires and surveys, our method offers an automated, and potentially more reliable way to assess these personality characteristics. Once these characteristics are considered, teams can be reshuffled and team membership changed for higher creativity.

Collaborative innovation networks, virtual communities and geographical clustering

Abstract: This paper describes the emergence of collaborative knowledge networks (CKNs), distributed communities taking advantage of the wide connectivity and the support of communication technologies, spanning beyond the organisational perimeter of companies on a global scale. CKNs are made up of groups of self-motivated individuals, linked by the idea of something new and exciting and by the common goal of improving existing business practices, new products or services for which they see a real need. Their strength is related to their ability to activate creative collaboration, knowledge sharing and social networking mechanisms, affecting positively individual capabilities and organisations' performance. We describe the case of a global consulting community to highlight the cultural and structural aspects of this phenomenon. Our case study also illustrates the composition of the CKN ecosystem, which are made up by a combination of collaborative innovation, learning and interest networks. Empirical evidence suggests physical proximity as a supporting success factor of such communities, depending on the capital and knowledge intensity of the target industry.

The Power of Alumni Networks - Success of Startup Companies Correlates With Online Social Network Structure of Its Founders

Abstract: In this paper we analyze the success of startups in Germany by looking at the social network structure of their founders on the German-language business-networking site XING. We address two related research questions. First we examine university-wide networks, constructing alumni networks of 12 German universities, with the goal of identifying the most successful founder networks among the 12 universities. Second, we also look at individual actor network structure, to find the social network attributes of the most successful founders. We automatically collected the publicly accessible portion of XING, filtering people by attributes indicative of their university, and roles as founders, entrepreneurs, and CEOs. We identified 51,976 alumni, out of which 14,854 have entrepreneurship attributes. We also manually evaluated the financial success of a subsample of 80 entrepreneurs for each university. We found that universities, which are more central in the German university network, provide a better environment for students to found more and more successful startups. University networks whose alumni have a stronger “old-boys-network”, i.e. a larger share of their links with other alumni of their alma mater, are more successful as founders of startups. On the individual level the same holds true: the more links founders have with alumni of their university, the more successful their startup is. Finally, the absolute amount of networking matters, i.e. the more links entrepreneurs have, and the higher their betweenness in the online network of university alumni, the more successful they are.

Process efficiency. Redesigning social networks to improve surgery patient flow.

Abstract: We propose a novel approach to improve throughput of the surgery patient flow process of a Boston area teaching hospital. A social network analysis was conducted in an effort to demonstrate that process efficiency gains could be achieved through redesign of social network patterns at the workplace; in conjunction with redesign of organization structure and the implementation of workflow over an integrated information technology system. Key knowledge experts and coordinators in times of crisis were identified and a new communication structure more conducive to trust and knowledge sharing was suggested. The new communication structure is scalable without compromising on coordination required among key roles in the network for achieving efficiency gains.

Does Geographic Clustering Still Benefit High Tech New Ventures? The Case of the Cambridge/Boston Biotech Cluster

Abstract: An empirical study of scientific communication among biotechnology companies supports the belief that geographic clustering does produce increased scientific exchange among companies. A comparison of companies within a constrained geographic area with those more dispersed shows a significantly higher level of scientific communication among the former. Scientific communication declines rapidly with plupical separation. Critical of the formation of cluster – based scientific communication networks is the presence of both universities and large firms from the same industry.

Analyse informeller Kommunikationsnetzwerke am Beispiel einer Fallstudie

Abstract: Die Struktur und Dynamik informeller Kommunikationsnetzwerke sind von zentraler Bedeutung fur das Funktionieren betrieblicher Arbeitsprozesse und beeinflussen die Leistungs- und Innovationsfahigkeit von wissensintensiven Organisationen. Wahrend sich die meisten Fuhrungskrafte dessen bewusst sind, fehlt es an (teil-)automatischen, IT-gestutzten Methoden und Instrumenten, die informelle Kommunikationsnetzwerke erfassbar machen. Wahrend die Protokollierung elektronisch vermittelter Kommunikation in den letzten Jahren deutliche Fortschritte gemacht hat, ist die Abbildung von personlichen Interaktionen nach wie vor sehr aufwandig und insbesondere bei manuellen Verfahren sehr fehleranfallig. Die Autoren zeigen in dem Beitrag, wie sich informelle Kommunikationsnetzwerke mit Hilfe von IT-gestutzten Verfahren untersuchen lassen. Dabei prasentieren sie ein Instrument („Social Badges“), das Daten uber die personliche Kommunikation automatisch und genauer erfasst, als dies mit herkommlichen Mitteln realisierbar ist. Die praktische Verwendbarkeit des Ansatzes wird anhand einer Fallstudie evaluiert.

Deriving taxonomies from automatic analysis of group membership structure in large social networks

Abstract: We develop a method to create taxonomies in large social networks solely based on users group membership information. We illustrate our technique using an example of the Flickr photo sharing network. This photo sharing community is a social network that enables users to collectively store digital pictures, interact and form groups of interest. Based on our earlier research of success indicators for individual actors on the Flickr Community, we extend the focus to groups of interest. Introducing the metric “GroupConnectivity” we perform a community segmentation. We then develop a method to automatically create taxonomies without the need for folksonomies or pre-existing ontologies.