Structural abstraction of process specifications

Software engineers constantly deal with problems of designing, analyzing, and improving process specifications, e.g., source code, service compositions, or process models. Process specifications are abstractions of behavior observed or intended to be implemented in reality which result from creative engineering practice. Usually, process specifications are formalized as directed graphs in which edges capture temporal relations between decisions, synchronization points, and work activities. Every process specification is a compromise between two points: On the one hand engineers strive to operate with less modeling constructs which conceal irrelevant details, while on the other hand the details are required to achieve the desired level of customization for envisioned process scenarios. In our research, we approach the problem of varying abstraction levels of process specifications. Formally, developed abstraction mechanisms exploit the structure of a process specification and allow the generalization of low-level details into concepts of a higher abstraction level. The reverse procedure can be addressed as process specialization.

Adolescent pedometer protocols : examining reactivity, tampering and participants' perceptions

The aim of this study was to investigate adolescents' potential reactivity and tampering while wearing pedometers by comparing different monitoring protocols to accelerometer output. The sample included adolescents (N=123, age range=14-15 years) from three secondary schools in New South Wales, Australia. Schools were randomised to one of the three pedometer monitoring protocols: (i) daily sealed (DS) pedometer group, (ii) unsealed (US) pedometer group or (iii) weekly sealed (WS) pedometer group. Participants wore pedometers (Yamax Digi-Walker CW700, Yamax Corporation, Kumamoto City, Japan) and accelerometers (Actigraph GT3X+, Pensacola, USA) simultaneously for seven days. Repeated measures analysis of variance was used to examine potential reactivity. Bivariate correlations between step counts and accelerometer output were calculated to explore potential tampering. The correlation between accelerometer output and pedometer steps/day was strongest among participants in the WS group (r=0.82, P <= 0.001), compared to the US (r=0.63, P <= 0.001) and DS (r=0.16, P=0.324) groups. The DS (P <= 0.001) and US (P=0.003), but not the WS (P=0.891), groups showed evidence of reactivity. The results suggest that reactivity and tampering does occur in adolescents and contrary to existing research, pedometer monitoring protocols may influence participant behaviour.

Investigating subspace distances in semantic spaces

Semantic space models of word meaning derived from co-occurrence statistics within a corpus of documents, such as the Hyperspace Analogous to Language (HAL) model, have been proposed in the past. While word similarity can be computed using these models, it is not clear how semantic spaces derived from different sets of documents can be compared. In this paper, we focus on this problem, and we revisit the proposal of using semantic subspace distance measurements [1]. In particular, we outline the research questions that still need to be addressed to investigate and validate these distance measures. Then, we describe our plans for future research.

Working notes for TopSig at ShARe/CLEF 2013

We used our TopSig open-source indexing and retrieval tool to produce runs for the ShARe/CLEF eHealth 2013 track. TopSig was used to produce runs using the query fields and provided discharge summaries, where appropriate. Although the improvement was not great TopSig was able to gain some benefit from utilising the discharge summaries, although the software needed to be modified to support this. This was part of a larger experiment involving determining the applicability and limits to signature-based approaches.

Ranking based clustering for social event detection

The problem of clustering a large document collection is not only challenged by the number of documents and the number of dimensions, but it is also affected by the number and sizes of the clusters. Traditional clustering methods fail to scale when they need to generate a large number of clusters. Furthermore, when the clusters size in the solution is heterogeneous, i.e. some of the clusters are large in size, the similarity measures tend to degrade. A ranking based clustering method is proposed to deal with these issues in the context of the Social Event Detection task. Ranking scores are used to select a small number of most relevant clusters in order to compare and place a document. Additionally,instead of conventional cluster centroids, cluster patches are proposed to represent clusters, that are hubs-like set of documents. Text, temporal, spatial and visual content information collected from the social event images is utilized in calculating similarity. Results show that these strategies allow us to have a balance between performance and accuracy of the clustering solution gained by the clustering method.

SAIVT-ADMRG @ MediaEval 2014 social event detection

This paper outlines the approach taken by the Speech, Audio, Image and Video Technologies laboratory, and the Applied Data Mining Research Group (SAIVT-ADMRG) in the 2014 MediaEval Social Event Detection (SED) task. We participated in the event based clustering subtask (subtask 1), and focused on investigating the incorporation of image features as another source of data to aid clustering. In particular, we developed a descriptor based around the use of super-pixel segmentation, that allows a low dimensional feature that incorporates both colour and texture information to be extracted and used within the popular bag-of-visual-words (BoVW) approach.

Fine-grained plant classification using convolutional neural networks for feature extraction

We present an overview of the QUT plant classification system submitted to LifeCLEF 2014. This system uses generic features extracted from a convolutional neural network previously used to perform general object classification. We examine the effectiveness of these features to perform plant classification when used in combination with an extremely randomised forest. Using this system, with minimal tuning, we obtained relatively good results with a score of 0:249 on the test set of LifeCLEF 2014.

Instance-driven TBox revision in DL-Lite

The development and maintenance of large and complex ontologies are often time-consuming and error-prone. Thus, automated ontology learning and revision have attracted intensive research interest. In data-centric applications where ontologies are designed or automatically learnt from the data, when new data instances are added that contradict to the ontology, it is often desirable to incrementally revise the ontology according to the added data. This problem can be intuitively formulated as the problem of revising a TBox by an ABox. In this paper we introduce a model-theoretic approach to such an ontology revision problem by using a novel alternative semantic characterisation of DL-Lite ontologies. We show some desired properties for our ontology revision. We have also developed an algorithm for reasoning with the ontology revision without computing the revision result. The algorithm is efficient as its computational complexity is in coNP in the worst case and in PTIME when the size of the new data is bounded.

A modified Density-Based Scanning Algorithm with Noise for spatial travel pattern analysis from Smart Card AFC data

Smart Card Automated Fare Collection (AFC) data has been extensively exploited to understand passenger behavior, passenger segment, trip purpose and improve transit planning through spatial travel pattern analysis. The literature has been evolving from simple to more sophisticated methods such as from aggregated to individual travel pattern analysis, and from stop-to-stop to flexible stop aggregation. However, the issue of high computing complexity has limited these methods in practical applications. This paper proposes a new algorithm named Weighted Stop Density Based Scanning Algorithm with Noise (WS-DBSCAN) based on the classical Density Based Scanning Algorithm with Noise (DBSCAN) algorithm to detect and update the daily changes in travel pattern. WS-DBSCAN converts the classical quadratic computation complexity DBSCAN to a problem of sub-quadratic complexity. The numerical experiment using the real AFC data in South East Queensland, Australia shows that the algorithm costs only 0.45% in computation time compared to the classical DBSCAN, but provides the same clustering results.

Deriving word association networks from text corpora

This article presents and evaluates a model to automatically derive word association networks from text corpora. Two aspects were evaluated: To what degree can corpus-based word association networks (CANs) approximate human word association networks with respect to (1) their ability to quantitatively predict word associations and (2) their structural network characteristics. Word association networks are the basis of the human mental lexicon. However, extracting such networks from human subjects is laborious, time consuming and thus necessarily limited in relation to the breadth of human vocabulary. Automatic derivation of word associations from text corpora would address these limitations. In both evaluations corpus-based processing provided vector representations for words. These representations were then employed to derive CANs using two measures: (1) the well known cosine metric, which is a symmetric measure, and (2) a new asymmetric measure computed from orthogonal vector projections. For both evaluations, the full set of 4068 free association networks (FANs) from the University of South Florida word association norms were used as baseline human data. Two corpus based models were benchmarked for comparison: a latent topic model and latent semantic analysis (LSA). We observed that CANs constructed using the asymmetric measure were slightly less effective than the topic model in quantitatively predicting free associates, and slightly better than LSA. The structural networks analysis revealed that CANs do approximate the FANs to an encouraging degree.

Implementing process views in the web service environment

Web service and business process technologies are widely adopted to facilitate business automation and collaboration. Given the complexity of business processes, it is a sought-after feature to show a business process with different views to cater for the diverse interests, authority levels, etc., of different users. Aiming to implement such flexible process views in the Web service environment, this paper presents a novel framework named FlexView to support view abstraction and concretisation of WS-BPEL processes. In the FlexView framework, a rigorous view model is proposed to specify the dependency and correlation between structural components of process views with emphasis on the characteristics of WS-BPEL, and a set of rules are defined to guarantee the structural consistency between process views during transformations. A set of algorithms are developed to shift the abstraction and concretisation operations to the operational level. A prototype is also implemented for the proof-of-concept purpose. © 2010 Springer Science+Business Media, LLC.

Non-finite constructions in Old English : with special reference to syntactic borrowing from Latin

My dissertation is a corpus-based study of non-finite constructions in Old English (OE). It revisits the question of Latin influence on the OE syntax, offering a new evaluation of syntactic interference between Latin and OE, and, more generally, of the contact situation in the OE period, drawing on methods used in studying grammaticalization and language contact. I address three non-finite constructions: absolute participial construction, accusative-and-infinitive construction, and nominative-and-infinitive construction, exemplified respectively in present-day English as - She looked like a pixie sometimes, her eyes darting here and there, forever watchful (BNC CCM 98); - My first acquaintance with her was when I heard her sing (BNC CFY 2215); - Charles the Bald was said to resemble his grandfather physically (BNC HPT 175). This study compares data from translated texts against the background of original OE writings, establishing dependencies and differences between the two. Although the contrastive analysis of source and target texts is one of the major methods employed in the study, translation and translation strategies as such are only my secondary foci. The emphasis is rather on what source/target comparison can tell us about the OE non-finite syntax and the typological differences between Latin and OE in this domain, and on whether contact-induced change can originate in translation. In terms of theoretical framework, I have adopted functional-typological approach, which rests on the principles of iconicity and event integration, and to the best of my knowledge, has not been applied systematically to OE non-finite constructions. Therefore one more aim of the dissertation is to test this framework and to see how OE fits into the cross-linguistic picture of non-finites. My research corpus consists of two samples: 1) written OE closely dependent on the Latin originals, based on editions of two gloss texts, five translations, and Latin originals of these texts, representing four text types: hymns, religious regulations, homily/life narrative, and biblical narrative (180,622 words); and 2) written OE as far independent from Latin as possible, based on a selection from the York-Toronto-Helsinki Parsed Corpus of Old English Prose (YCOE) and representing five text types: laws, charters, correspondence, chronicle narrative, and homily/life narrative (274,757 words).

Biological control of parthenium (Parthenium hysterophorus) in Australian rangeland translates to improved grass production.

Biological control of parthenium, a major weed in grazing areas in Australia, was initiated in the mid 1970s. Since then, nine species of insects and two rust fungi have been introduced. Evaluation using pesticide exclusion at two sites (Mt. Panorama and Plain Creek) in Queensland, Australia, revealed that classical biological control had a significant negative effect on the target weed, but the impact varied between years. In this study, I quantified the effects of biological control of parthenium on grass production. Grass production declined with the increase in parthenium biomass. Significant increase in grass production due to biological control was observed, but only in 1 of 4 yr at Mt. Panorama and 2 of 4 yr at Plain Creek. At Mt. Panorama, there was a 40% increase in grass biomass in 1997 because of defoliation by Zygogramma bicolorata and galling by Epiblema strenuana. At Plain Creek, grass biomass increased by 52% in 1998 because of E. strenuana and by 45% in 2000 because of combined effects of E. strenuana and the summer rust Puccinia melampodii. This study provides evidence on the beneficial effects of biological control of parthenium in areas under limited grazing.

Evaluation of the Australian branched broomrape (Orobanche ramosa) eradication program

Because weed eradication programs commonly take 10 or more years to complete, there is a need to evaluate progress toward the eradication objective. We present a simple model, based on information that is readily obtainable, that assesses conformity to the delimitation and extirpation criteria for eradication. It is applied to the program currently targeting the annual parasitic weed, branched broomrape, in South Australia. The model consists of delimitation and extirpation (E) measures plotted against each other to form an 'eradograph.' Deviations from the 'ideal' eradograph plot can inform tactical responses, e.g., increases in survey and/or control effort. Infestations progress from the active phase to the monitoring phase when no plants have been detected for at least 12 mo. They revert to the active phase upon further detection of plants. We summarize this process for the invasion as a whole in a state-and-transition model. Using this model we demonstrate that the invasion is unlikely to be delimited unless the amount of newly detected infested area decreases, on average, by at least 50% per annum. As a result of control activities implemented, on average approximately 70% (range, 44 to 86%) of active infestations progressed to the monitoring phase in the year following their detection. Simulations suggest that increasing this rate of transition will not increase E to a significant extent. The rate of reversion of infestations from the monitoring phase to the active phase decreased logarithmically with time since last detection, but it is likely that lower rates of reversion would accelerate the trend toward extirpation. Program performance with respect to the delimitation criterion has been variable; performance with respect to the extirpation criterion would be improved considerably by the development and application of cost-effective methods for eliminating branched broomrape soil seed populations.