Human activity learning and segmentation using partially hidden discriminative models

Learning and understanding the typical patterns in the daily activities and routines of people from low-level sensory data is an important problem in many application domains such as building smart environments, or providing intelligent assistance. Traditional approaches to this problem typically rely on supervised learning and generative models such as the hidden Markov models and its extensions. While activity data can be readily acquired from pervasive sensors, e.g. in smart environments, providing manual labels to support supervised training is often extremely expensive. In this paper, we propose a new approach based on semi-supervised training of partially hidden discriminative models such as the conditional random field (CRF) and the maximum entropy Markov model (MEMM). We show that these models allow us to incorporate both labeled and unlabeled data for learning, and at the same time, provide us with the flexibility and accuracy of the discriminative framework. Our experimental results in the video surveillance domain illustrate that these models can perform better than their generative counterpart, the partially hidden Markov model, even when a substantial amount of labels are unavailable.

Symbolic representation and distributed matching strategies for schematics

This paper describes object-centered symbolic representation and distributed matching strategies of 3D objects in a schematic form which occur in engineering drawings and maps. The object-centered representation has a hierarchical structure and is constructed from symbolic representations of schematics. With this representation, two independent schematics representing the same object can be matched. We also consider matching strategies using distributed algorithms. The object recognition is carried out with two matching methods: (1) matching between an object model and observed data at the lowest level of the hierarchy, and (2) constraints propagation. The first is carried out with symbolic Hopfield-type neural networks and the second is achieved via hierarchical winner-takes-all algorithms

Strategic thinking and the creative process in event management

This thesis asserts that event management requires strategic and creative thinking. This is necessary to envision alternative approaches in an evolving communications landscape. Using a rich, qualitative analysis with one major case study, the conclusion is that events must shift from an operationally led to a strategically informed creative process.

Symbolic interactions and the development of women leaders in higher education

The under-representation of women in higher education (HE) leadership is a persistent global phenomenon. The purpose of this research is to re-examine this issue through symbolic interactionism (SI). Eight women aspiring to leadership were invited to participate in semi-structured interviews after attending a leadership programme specifically designed to enhance their leadership prospects. Analysis indicated ambiguities and contradictions that surround notions of leadership, in particular how the participants position themselves and are positioned in their workplace. This was evidenced by the meaning they attributed to: recognizing a leader; interactions with existing leadership; and speculation regarding their leadership capacity. Gendered notions were apparent in their constructed meanings. Similarity attraction was also evident, with men being observed as ‘paying it forward’, therefore facilitating promotion. Formal leadership training was advocated rather than experiential processes.

Counterfactual thinking and anticipated emotions enhance performance in computer skills training

The present study examined the relationship between novice learners’ counterfactual thinking (i.e. generating ‘what if’ and ‘if only’ thoughts) about their initial training experience with a computer application and subsequent improvement in task performance. The role of anticipated emotions towards goal attainment in task performance was also assessed. Undergraduate students (N ¼ 42) with minimal experience in using computer spreadsheets underwent basic training in using Microsoft Excel. All participants were assessed on their anticipated positive and negative emotions regarding goal attainment at the outset. After completing their first task, participants allocated to a counterfactual condition received instructions to generate counterfactual thoughts regarding their initial task performance, whereas participants in a control condition did not. The counterfactual group showed only marginally greater improvement in task performance (measured by task completion time and accuracy) than the control group.  However, we also found that positive anticipated emotions were associated with improvement in task performance but for the counterfactual group only. Our data have implications for incorporating counterfactual thinking into information technology skills training to enhance learning outcomes for novice learners.