CSFinder: A Cold-Start Friend Finder in Large-Scale Social Networks

Recommending users for a new social network user to follow is a topic of interest at present. The existing approaches rely on using various types of information about the new user to determine recommended users who have similar interests to the new user. However, this presents a problem when a new user joins a social network, who is yet to have any interaction on the social network. In this paper we present a particular type of conversational recommendation approach, critiquing-based recommendation, to solve the cold start problem. We present a critiquing-based recommendation system, called CSFinder, to recommend users for a new user to follow. A traditional critiquing-based recommendation system allows a user to critique a feature of a recommended item at a time and gradually leads the user to the target recommendation. However this may require a lengthy recommendation session. CSFinder aims to reduce the session length by taking a case-based reasoning approach. It selects relevant recommendation sessions of past users that match the recommendation session of the current user to shortcut the current recommendation session. It selects relevant recommendation sessions from a case base that contains the successful recommendation sessions of past users. A past recommendation session can be selected if it contains recommended items and critiques that sufficiently overlap with the ones in the current session. Our experimental results show that CSFinder has significantly shorter sessions than the ones of an Incremental Critiquing system, which is a baseline critiquing-based recommendation system.

Two-part Segmentation of Text Documents

We consider the problem of segmenting text documents that have a
two-part structure such as a problem part and a solution part. Documents
of this genre include incident reports that typically involve
description of events relating to a problem followed by those pertaining
to the solution that was tried. Segmenting such documents
into the component two parts would render them usable in knowledge
reuse frameworks such as Case-Based Reasoning. This segmentation
problem presents a hard case for traditional text segmentation
due to the lexical inter-relatedness of the segments. We develop
a two-part segmentation technique that can harness a corpus
of similar documents to model the behavior of the two segments
and their inter-relatedness using language models and translation
models respectively. In particular, we use separate language models
for the problem and solution segment types, whereas the interrelatedness
between segment types is modeled using an IBM Model
1 translation model. We model documents as being generated starting
from the problem part that comprises of words sampled from
the problem language model, followed by the solution part whose
words are sampled either from the solution language model or from
a translation model conditioned on the words already chosen in the
problem part. We show, through an extensive set of experiments on
real-world data, that our approach outperforms the state-of-the-art
text segmentation algorithms in the accuracy of segmentation, and
that such improved accuracy translates well to improved usability
in Case-based Reasoning systems. We also analyze the robustness
of our technique to varying amounts and types of noise and empirically
illustrate that our technique is quite noise tolerant, and
degrades gracefully with increasing amounts of noise

Knowledge Management Techniques to Augment Textual CBR (PhD Thesis)

The past decade had witnessed an unprecedented growth in the amount of available digital content, and its volume is expected to continue to grow the next few years. Unstructured text data generated from web and enterprise sources form a large fraction of such content. Many of these contain large volumes of reusable data such as solutions to frequently occurring problems, and general know-how that may be reused in appropriate contexts. In this work, we address issues around leveraging unstructured text data from sources as diverse as the web and the enterprise within the Case-based Reasoning framework. Case-based Reasoning (CBR) provides a framework and methodology for systematic reuse of historical knowledge that is available in the form of problemsolution
pairs, in solving new problems. Here, we consider possibilities of enhancing Textual CBR systems under three main themes: procurement, maintenance and retrieval. We adapt and build upon the stateof-the-art techniques from data mining and natural language processing in addressing various challenges therein. Under procurement, we investigate the problem of extracting cases (i.e., problem-solution pairs) from data sources such as incident/experience
reports. We develop case-base maintenance methods specifically tuned to text targeted towards retaining solutions such that the utility of the filtered case base in solving new problems is maximized. Further, we address the problem of query suggestions for textual case-bases and show that exploiting the problem-solution partition can enhance retrieval effectiveness by prioritizing more useful query suggestions. Additionally, we illustrate interpretable clustering as a tool to drill-down to domain specific text collections (since CBR systems are usually very domain specific) and develop techniques for improved similarity assessment in social media sources such as microblogs. Through extensive empirical evaluations, we illustrate the improvements that we are able to
achieve over the state-of-the-art methods for the respective tasks.

Integrating textual analysis and evidential reasoning for decision making in Engineering design

Decision making is an important element throughout the life-cycle of large-scale projects. Decisions are critical as they have a direct impact upon the success/outcome of a project and are affected by many factors including the certainty and precision of information. In this paper we present an evidential reasoning framework which applies Dempster-Shafer Theory and its variant Dezert-Smarandache Theory to aid decision makers in making decisions where the knowledge available may be imprecise, conflicting and uncertain. This conceptual framework is novel as natural language based information extraction techniques are utilized in the extraction and estimation of beliefs from diverse textual information sources, rather than assuming these estimations as already given. Furthermore we describe an algorithm to define a set of maximal consistent subsets before fusion occurs in the reasoning framework. This is important as inconsistencies between subsets may produce results which are incorrect/adverse in the decision making process. The proposed framework can be applied to problems involving material selection and a Use Case based in the Engineering domain is presented to illustrate the approach. © 2013 Elsevier B.V. All rights reserved.

How many processes underline category-based induction? Effect of conclusion specificity and cognitive ability

Two studies investigated participants' sensitivity to the amount and diversity of the evidence when reasoning inductively about categories. Both showed that participants are more sensitive to characteristics of the evidence for arguments with general rather than specific conclusions. Both showed an association between cognitive ability and sensitivity to these evidence characteristics, particularly when the conclusion category was general. These results suggest that a simple associative process may not be sufficient to capture some key phenomena of category-based induction. They also support the claim that the need to generate a superordinate category is a complicating factor in category-based reasoning and that adults' tendency to generate such categories while reasoning has been overestimated.

Educational interventions in neurology:A comprehensive systematic review

A fear of neurology and neural sciences (neurophobia) may have clinical consequences. There is therefore a need to formulate an evidence-based approach to neurology education. A comprehensive systematic review of educational interventions in neurology was performed. BEI, Cochrane Library, Dialog Datastar, EBSCO Biomedical, EBSCO Psychology & Behavioral Sciences, EMBASE, ERIC, First Search, MDConsult, Medline, Proquest Medical Library and Web of Knowledge databases were searched for all published studies assessing interventions in neurology education among undergraduate students, junior medical doctors and residents up to and including July 2012. Two independent literature searches were performed for relevant studies, which were then classified for level of evidence using the Centre of Evidence-based Medicine criteria and four levels of Kirkpatrick educational outcomes. One systematic review, 16 randomized controlled trials (RCTs), nine non-randomized cohort/follow-up studies, 33 case series or historically controlled studies and three mechanism-based reasoning studies were identified. Educational interventions showed favourable evaluation or assessment outcomes in 15 of 16 (94%) RCTs. Very few studies measured subsequent clinical behaviour (two studies) and patient outcomes (one study). There is very little high quality evidence of demonstrably effective neurology education. However, RCTs are emerging, albeit without meeting comprehensive educational criteria. An improving evidence base in the quality of neurology education will be important to reduce neurophobia. © 2013 EFNS.

Parallel Time Series Analysis and Income Inequality: A New Approach to Comparative Case Studies

This case study deals with the role of time series analysis in sociology, and its relationship with the wider literature and methodology of comparative case study research. Time series analysis is now well-represented in top-ranked sociology journals, often in the form of ‘pooled time series’ research designs. These studies typically pool multiple countries together into a pooled time series cross-section panel, in order to provide a larger sample for more robust and comprehensive analysis. This approach is well suited to exploring trans-national phenomena, and for elaborating useful macro-level theories specific to social structures, national policies, and long-term historical processes. It is less suited however, to understanding how these global social processes work in different countries. As such, the complexities of individual countries - which often display very different or contradictory dynamics than those suggested in pooled studies – are subsumed. Meanwhile, a robust literature on comparative case-based methods exists in the social sciences, where researchers focus on differences between cases, and the complex ways in which they co-evolve or diverge over time. A good example of this is the inequality literature, where although panel studies suggest a general trend of rising inequality driven by the weakening power of labour, marketisation of welfare, and the rising power of capital, some countries have still managed to remain resilient. This case study takes a closer look at what can be learned by applying the insights of case-based comparative research to the method of time series analysis. Taking international income inequality as its point of departure, it argues that we have much to learn about the viability of different combinations of policy options by examining how they work in different countries over time. By taking representative cases from different welfare systems (liberal, social democratic, corporatist, or antipodean), we can better sharpen our theories of how policies can be more specifically engineered to offset rising inequality. This involves a fundamental realignment of the strategy of time series analysis, grounding it instead in a qualitative appreciation of the historical context of cases, as a basis for comparing effects between different countries.

A programming model for BSP with partitioned synchronisation

A BSP superstep is a distributed computation comprising a number of simultaneously executing processes which may generate asynchronous messages. A superstep terminates with a barrier which enforces a global synchronisation and delivers all ongoing communications. Multilevel supersteps can utilise barriers in which subsets of processes, interacting through shared memories, are locally synchronised (partitioned synchronisation). In this paper a state-based semantics, closely related to the classical sequential programming model, is derived for distributed BSP with partitioned synchronisation.

An exploration of the attitudes and strategic responses of soleproprietor micro-enterprises in adopting information and communication technology

This article analyses longitudinal case-based research exploring the attitudes and strategic responses of micro-enterprise owners in adopting information and communication technology (ICT). In so doing, it contributes to the limited literature on micro-enterprise ICT adoption, with a particular focus on sole proprietors. It provides a basis for widening the theoretical base of the literature pertaining to ICT adoption on two levels. First, a framework is developed which integrates the findings to illustrate the relationships between attitudes towards ICT adoption, endogenous and exogenous influencers of these attitudes and subsequent strategic response in ICT adoption. Second, building upon this framework the article reveals the unique challenges, opportunities and implications of ICT adoption for sole-proprietor micro-enterprises. © The Author(s) 2012

Digital loyalty card 'Big Data' and small business marketing: Formal versus informal or complementary?

This article proposes that a complementary relationship exists between the formalised nature of digital loyalty card data, and the informal nature of small business market orientation. A longitudinal, case-based research approach analysed this relationship in small firms given access to Tesco Clubcard data. The findings reveal a new-found structure and precision in small firm marketing planning from data exposure; this complemented rather than conflicted with an intuitive feel for markets. In addition, small firm owners were encouraged to include employees in marketing planning.

An exploration of business model development in the commercialization of technology innovations

Research on business model development has focused on the relationships between elements of value conceptualization and organization having a linear sequence in which business models are first designed and then implemented. Another stream of research points to business model development with these elements interacting in a cyclical manner. There is a need to improve our understanding of the connective mechanisms and dynamics involved in business model development, particularly from the challenging perspective of commercializing innovations. The aim of this paper was to explore business model development during the commercialization of innovations through a case-based qualitative study. This study found from four case studies that specific elements of business model development, representative of the conceptualization of value and organizing for value creation, integrate in a dynamic and cyclical process in the commercialization of technology innovations. The study provides empirical evidence that adds new insights to literature on sequential and more interactive processes of business model development. It also contributes to literature on business model development and particularly how it relates to the commercialization of innovations.