Basics of qualitative research: Grounded theory procedures and techniques

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

This paper presents an overview of the field of recommender systems and describes the current generation of recommendation methods that are usually classified into the following three main categories: content-based, collaborative, and hybrid recommendation approaches. This paper also describes various limitations of current recommendation methods and discusses possible extensions that can improve recommendation capabilities and make recommender systems applicable to an even broader range of applications. These extensions include, among others, an improvement of understanding of users and items, incorporation of the contextual information into the recommendation process, support for multicriteria ratings, and a provision of more flexible and less intrusive types of recommendations.

/pdf/toward-the-next-generation-of-recommender-systems-a-survey-4wwsv77rns.pdf

Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions

Data Mining Practical Machine Learning Tools and Techniques

How has Japan become a major economic power, a world leader in the automotive and electronics industries? What is the secret of their success? The consensus has been that, though the Japanese are not particularly innovative, they are exceptionally skilful at imitation, at improving products that already exist. But now two leading Japanese business experts, Ikujiro Nonaka and Hiro Takeuchi, turn this conventional wisdom on its head: Japanese firms are successful, they contend, precisely because they are innovative, because they create new knowledge and use it to produce successful products and technologies. Examining case studies drawn from such firms as Honda, Canon, Matsushita, NEC, 3M, GE, and the U.S. Marines, this book reveals how Japanese companies translate tacit to explicit knowledge and use it to produce new processes, products, and services.

The knowledge-creating company : how Japanese companies create the dynamics of innovation

A number of studies in the network science literature have attempted to model the effect of network structure on cognitive state fluctuations in social networks. For the most part, these networks use highly simplified models of both cognitive state and social influence. In order to extend these studies and provide the basis for more complex network science simulations, a model of socially-mediated cognitive change is presented. The model attempts to integrate ideas and concepts from a number of disciplines, most notably psychology, evolutionary biology and complexity science. In the model, cognitive states are modelled as networks of binary variables, each of which indicates an agent’s belief in a particular fact. The links between variables represent the ‘logical’ dependencies between beliefs, and these dependencies are based on an agent’s knowledge of the domain to which the beliefs apply. Drawing on the psychological notion of cognitive dissonance, it is further suggested that agents are under internal pressure to adopt highly consistent belief configurations, and this identifies one source of cognitive dynamism in the model. Another source of dynamism derives from the structure of the social network. Here, the existence of network ties creates a dependency between the belief systems of connected agents. Cognitive change in such ‘coupled belief systems’ is modelled using Kauffman’s NK(C) model of co-evolutionary development in biological systems. As a final source of cognitive dynamism, the model incorporates the notion of an aggregate belief system (or cultural model), which represents the dominant set of beliefs associated with specific agent sub-groups. By explicitly incorporating the notion of an aggregate belief system into the model, the model supports the analysis of cognitive state fluctuations at the individual (psychological), social and cultural levels. It also provides the basis for future network science simulations that seek to study the complex interactions between these various levels.

Modelling the Dynamics of Collective Cognition: A Network-Based Approach to Socially-Mediated Cognitive Change

Social Network Analysis (SNA) and Web Mining (WM) techniques are being applied to study the structures of social networks in order to manage their dynamics and predict their evolution. This paper describes how we used Semantically-Interlinked Online Communities (SIOC) ontology to represent the (latent) semantic relationships between the members of a large community forum (about 2,500), Plexilandia. We extended SIOC, taking advantage of topic-based text mining and developed data mining algorithms that used our SIOC extensions to provide a better understanding of the social dynamics of the members of the Plexilandia community. This new understanding helped us to detect and discover the key members of Plexilandia successfuly.

/pdf/leveraging-social-network-analysis-with-topic-models-and-the-2gcuea9jd9.pdf

Leveraging Social Network Analysis with Topic Models and the Semantic Web

• We analyse ratings & reviews for 334 digital self-control tools from three app stores. • The main use is to focus on less instantly rewarding tasks amidst digital distraction. • Users seek tools that can adapt to their personal definitions of distraction. • Users want support that is ‘just enough’ to change behaviour without feeling coercive. • Tools combining design patterns (e.g., blocking + goal reminders) have higher ratings. Much effort has been invested in designing digital systems that keep people ‘hooked’. By contrast, comparatively little is known about how designers can support people in re-gaining control. Online, however, hundreds of apps and browser extensions promise to help people self-regulate use of digital devices. Reviews and popularity metrics for these digital self-control tools (DSCTs) can indicate which design patterns are useful in the wild. Moreover, they reveal how platforms like Android and iOS differ in the ecosystems they enable for DSCTs, which has important implications for end users. We analysed reviews, installation numbers, and ratings for 334 DSCTs on the Google Play, Chrome Web, and Apple App stores, investigating what user reviews reveal about usage contexts and key design challenges, and how functionality relates to popularity metrics. Our thematic analysis of 1,529 reviews (sampled from a data set of 53,978 distinct reviews scraped in March 2019) found that DSCTs are seen as highly important for focusing on less instantly rewarding tasks when digital distractions are easily available. Users seek DSCTs that adapt to their personal definitions of distraction, and provide support that is sufficient to change behaviour without feeling too coercive. Reviewers suggested combining design patterns to provide a level of support that is ‘just right’. This was mirrored in the ratings, where tools that combine different types of design patterns (e.g., website blocking and goal reminders) tended to receive higher ratings than those implementing a single type. We discuss implications for research and design, including how design patterns in DSCTs interact, and how psychological reactance to DSCTs can be reduced. 

/pdf/the-goldilocks-level-of-support-using-user-reviews-ratings-28az4txt.pdf

The Goldilocks level of support: Using user reviews, ratings, and installation numbers to investigate digital self-control tools

The design and maintenance of complex engineering systems such as a jet engine generates a significant amount of documentation. Increasingly, aerospace manufacturers are shifting their focus from selling products to providing services. As a result, when designing new products, engineers must increasingly consider the engine's complete life-cycle as part of the design process. To identify possible areas of concern, engineers must obtain knowledge gained from the entire life of similar engines. However, because of the size and distributed nature of the company's operation, engineers often do not have access to front-line maintenance data. In addition, the large number of documents accrued makes it impossible for them to be examined thoroughly. This paper presents a prototype knowledge-based document repository for such an application. It searches and analyzes distributed document resources, and provides engineers with a summary view of the underlying knowledge. The aim is to allow engineers to incorporate maintenance issues into the initial design. Unlike existing document repositories and digital libraries, our approach is knowledge-based, where users browse summary reports instead of following suggested links. To test the validity of our proposed architecture, we have developed and deployed a working prototype.

/pdf/knowledge-management-for-a-large-service-oriented-22abx58x11.pdf

Knowledge management for a large service-oriented corporation

The talk reviews the history of development of Open Government Data with particular reference to the UK effort and data.gov.uk - it discusses the motivations and reasons why Governments, communities and individuals support Open Data initiatives. It presents a range of challenges that this throws up from cultural to organisational, information assurance to revenue models.

Nigel Shadbolt

Papers

Modelling the Dynamics of Collective Cognition: A Network-Based Approach to Socially-Mediated Cognitive Change

Leveraging Social Network Analysis with Topic Models and the Semantic Web

The Goldilocks level of support: Using user reviews, ratings, and installation numbers to investigate digital self-control tools

Knowledge management for a large service-oriented corporation

Why Open Government Data? ?Lessons from data.gov.uk