Digital archives and cultural memory : discovering lost histories in digitised Australian children's literature 1851-1945

A recent Australian literature digitisation project uncovered some surprising discoveries in the children’s books that it digitised. The Children’s Literature Digital Resources (CLDR) Project digitised children’s books that were first published between 1851 to 1945 and made them available online through AustLit: The Australian Literature Resource. The digitisation process also preserved, within the pages of those books, a range of bookplates, book labels, inscriptions, and loose ephemera. This material allows us to trace the provenance of some of the digitised works, some of which came from the personal libraries of now-famous authors, and others from less celebrated sources. These extra-textual traces can contribute to cultural memory of the past by providing evidence of how books were collected and exchanged, and what kinds of books were presented as prizes in schools and Sunday schools. They also provide insight into Australian literary and artistic networks, particularly of the first few decades of the 20th century. This article describes the kinds of material uncovered in the digitisation process and suggests that the material provides insights into literary and cultural histories that might otherwise be forgotten. It also argues that the indexing of this material is vital if it is not to be lost to future researchers.

Transforming mobile personal life log into autobiographical multimedia eChronicles

Mobile devices are becoming indispensable personal assistants in people's daily life as these devices support work, study, play and socializing activities. The multi-modal sensors and rich features of smartphones can capture abundant information about users' life experience, such as taking photos or videos on what they see and hear, and organizing their tasks and activities using calendar, to-do lists, and notes. Such vast information can become useful to help users recalling episodic memories and reminisce about meaningful experiences. In this paper, we propose to apply autobiographical memory framework to provide an effective mechanism to structure mobile life-log data. The proposed model is an attempt towards a more complete personal life-log indexing model, which will support long term capture, organization, and retrieval. To demonstrate the benefits of the proposed model, we propose some design solutions for enabling users-driven capture, annotation, and retrieval of autobiographical multimedia chronicles tools.

Speaker diarization : "who spoke when"

Speaker diarization is the process of annotating an input audio with information that attributes temporal regions of the audio signal to their respective sources, which may include both speech and non-speech events. For speech regions, the diarization system also specifies the locations of speaker boundaries and assign relative speaker labels to each homogeneous segment of speech. In short, speaker diarization systems effectively answer the question of ‘who spoke when’. There are several important applications for speaker diarization technology, such as facilitating speaker indexing systems to allow users to directly access the relevant segments of interest within a given audio, and assisting with other downstream processes such as summarizing and parsing. When combined with automatic speech recognition (ASR) systems, the metadata extracted from a speaker diarization system can provide complementary information for ASR transcripts including the location of speaker turns and relative speaker segment labels, making the transcripts more readable. Speaker diarization output can also be used to localize the instances of specific speakers to pool data for model adaptation, which in turn boosts transcription accuracies. Speaker diarization therefore plays an important role as a preliminary step in automatic transcription of audio data. The aim of this work is to improve the usefulness and practicality of speaker diarization technology, through the reduction of diarization error rates. In particular, this research is focused on the segmentation and clustering stages within a diarization system. Although particular emphasis is placed on the broadcast news audio domain and systems developed throughout this work are also trained and tested on broadcast news data, the techniques proposed in this dissertation are also applicable to other domains including telephone conversations and meetings audio. Three main research themes were pursued: heuristic rules for speaker segmentation, modelling uncertainty in speaker model estimates, and modelling uncertainty in eigenvoice speaker modelling. The use of heuristic approaches for the speaker segmentation task was first investigated, with emphasis placed on minimizing missed boundary detections. A set of heuristic rules was proposed, to govern the detection and heuristic selection of candidate speaker segment boundaries. A second pass, using the same heuristic algorithm with a smaller window, was also proposed with the aim of improving detection of boundaries around short speaker segments. Compared to single threshold based methods, the proposed heuristic approach was shown to provide improved segmentation performance, leading to a reduction in the overall diarization error rate. Methods to model the uncertainty in speaker model estimates were developed, to address the difficulties associated with making segmentation and clustering decisions with limited data in the speaker segments. The Bayes factor, derived specifically for multivariate Gaussian speaker modelling, was introduced to account for the uncertainty of the speaker model estimates. The use of the Bayes factor also enabled the incorporation of prior information regarding the audio to aid segmentation and clustering decisions. The idea of modelling uncertainty in speaker model estimates was also extended to the eigenvoice speaker modelling framework for the speaker clustering task. Building on the application of Bayesian approaches to the speaker diarization problem, the proposed approach takes into account the uncertainty associated with the explicit estimation of the speaker factors. The proposed decision criteria, based on Bayesian theory, was shown to generally outperform their non- Bayesian counterparts.

Robust automatic speaker linking and attribution

This research makes a major contribution which enables efficient searching and indexing of large archives of spoken audio based on speaker identity. It introduces a novel technique dubbed as “speaker attribution” which is the task of automatically determining ‘who spoke when?’ in recordings and then automatically linking the unique speaker identities within each recording across multiple recordings. The outcome of the research will also have significant impact in improving the performance of automatic speech recognition systems through the extracted speaker identities.

Some extensions to representation and encoding of structure in models of distributional semantics

This thesis makes several contributions towards improved methods for encoding structure in computational models of word meaning. New methods are proposed and evaluated which address the requirement of being able to easily encode linguistic structural features within a computational representation while retaining the ability to scale to large volumes of textual data. Various methods are implemented and evaluated on a range of evaluation tasks to demonstrate the effectiveness of the proposed methods.

Knowledge-based urban development performance analysis : moving towards an integrated system

Knowledge-based urban development (KBUD) has become the new development paradigm for the cities of the global knowledge economy era. Nevertheless, to date international KBUD performance analysis of prosperous knowledge cities is understudied. This paper, therefore, introduces the methodology and application of a novel performance analysis approach to comprehensively scrutinise the global perspectives on KBUD of cities—i.e., The KBUD Assessment Model (KBUD/AM). This indexing model puts 11 renowned knowledge cities—i.e., Birmingham, Boston, Brisbane, Helsinki, Istanbul, Manchester, Melbourne, San Francisco, Sydney, Toronto, Vancouver—under the KBUD microscope to provide a benchmarked international outlook. The results of the indexing provide internationally benchmarked snapshot of the degree of achievements in various KBUD performance areas. This paper discusses the further development avenues and potentialities of the index to become an integrated system for the policy-making circles of cities to benchmark themselves against their competitors and develop relevant KBUD policies.

A novel representation of bioacoustic events for content-based search in field audio data

Bioacoustic data can provide an important base for environmental monitoring. To explore a large amount of field recordings collected, an automated similarity search algorithm is presented in this paper. A region of an audio defined by frequency and time bounds is provided by a user; the content of the region is used to construct a query. In the retrieving process, our algorithm will automatically scan through recordings to search for similar regions. In detail, we present a feature extraction approach based on the visual content of vocalisations – in this case ridges, and develop a generic regional representation of vocalisations for indexing. Our feature extraction method works best for bird vocalisations showing ridge characteristics. The regional representation method allows the content of an arbitrary region of a continuous recording to be described in a compressed format.

Efficient top-k retrieval with signatures

This paper describes a new method of indexing and searching large binary signature collections to efficiently find similar signatures, addressing the scalability problem in signature search. Signatures offer efficient computation with acceptable measure of similarity in numerous applications. However, performing a complete search with a given search argument (a signature) requires a Hamming distance calculation against every signature in the collection. This quickly becomes excessive when dealing with large collections, presenting issues of scalability that limit their applicability. Our method efficiently finds similar signatures in very large collections, trading memory use and precision for greatly improved search speed. Experimental results demonstrate that our approach is capable of finding a set of nearest signatures to a given search argument with a high degree of speed and fidelity.

A parcel-scale assessment tool to measure sustainability through urban ecosystem components : the MUSIX model

In recent years a number of urban sustainability assessment frameworks are developed to better inform policy formulation and decision-making processes. This paper introduces one of these attempts in developing a comprehensive assessment tool—i.e., Micro-level Urban-ecosystem Sustainability IndeX (MUSIX). Being an indicator-based indexing model, MUSIX investigates the environmental impacts of land-uses on urban sustainability by measuring urban ecosystem components in local scale. The paper presents the methodology of MUSIX and demonstrates the performance of the model in a pilot test-bed—i.e., in Gold Coast, Australia. The model provides useful insights on the sustainability performance of the test-bed area. The parcel-scale findings of the indicators are used to identify local problems considering six main issues of urban development—i.e., hydrology; ecology; pollution; location; design, and; efficiency. The composite index score is used to propose betterment strategies to guide the development of local area plans in conjunction with the City's Planning Scheme. In overall, this study has shown that parcel-scale environmental data provides an overview of the local sustainability in urban areas as in the example of Gold Coast, which can also be used for setting environmental policy, objectives and targets.

Management of multiple heterogeneous UAVs using capability and autonomy visualisation: theory, experiment and result

The interest in utilising multiple heterogeneous Unmanned Aerial Vehicles (UAVs) in close proximity is growing rapidly. As such, many challenges are presented in the effective coordination and management of these UAVs; converting the current n-to-1 paradigm (n operators operating a single UAV) to the 1-to-n paradigm (one operator managing n UAVs). This paper introduces an Information Abstraction methodology used to produce the functional capability framework initially proposed by Chen et al. and its Level Of Detail (LOD) indexing scale. This framework was validated through comparing the operator workload and Situation Awareness (SA) of three experiment scenarios involving multiple autonomously heterogeneous UAVs. The first scenario was set in a high LOD configuration with highly abstracted UAV functional information; the second scenario was set in a mixed LOD configuration; and the final scenario was set in a low LOD configuration with maximal UAV functional information. Results show that there is a significant statistical decrease in operator workload when a UAV’s functional information is displayed at its physical form (low LOD - maximal information) when comparing to the mixed LOD configuration.

Topic dependent language modelling for spoken term detection

This paper investigates the effect of topic dependent language models (TDLM) on phonetic spoken term detection (STD) using dynamic match lattice spotting (DMLS). Phonetic STD consists of two steps: indexing and search. The accuracy of indexing audio segments into phone sequences using phone recognition methods directly affects the accuracy of the final STD system. If the topic of a document in known, recognizing the spoken words and indexing them to an intermediate representation is an easier task and consequently, detecting a search word in it will be more accurate and robust. In this paper, we propose the use of TDLMs in the indexing stage to improve the accuracy of STD in situations where the topic of the audio document is known in advance. It is shown that using TDLMs instead of the traditional general language model (GLM) improves STD performance according to figure of merit (FOM) criteria.

Phonetic spoken term search using topic information

The aim of spoken term detection (STD) is to find all occurrences of a specified query term in a large audio database. This process is usually divided into two steps: indexing and search. In a previous study, it was shown that knowing the topic of an audio document would help to improve the accuracy of indexing step which results in a better performance for STD system. In this paper, we propose the use of topic information not only in the indexing step, but also in the search step. Results of our experiments show that topic information could also be used in search step to improve the STD accuracy.

Economic Analysis and Policy moves forward

As Editor of Economic Analysis and Policy (EAP) I am delighted to announce that EAP is now published by Elsevier. EAP is the journal of the Economic Society of Australia (Queensland branch). As a result of this move, four issues of EAP will be published per year instead of the current three. This will include special issues. EAP will now receive wider coverage in the relevant abstracting and indexing services...

Explanation of relevance judgement discrepancy with quantum interference

A key concept in many Information Retrieval (IR) tasks, e.g. document indexing, query language modelling, aspect and diversity retrieval, is the relevance measurement of topics, i.e. to what extent an information object (e.g. a document or a query) is about the topics. This paper investigates the interference of relevance measurement of a topic caused by another topic. For example, consider that two user groups are required to judge whether a topic q is relevant to a document d, and q is presented together with another topic (referred to as a companion topic). If different companion topics are used for different groups, interestingly different relevance probabilities of q given d can be reached. In this paper, we present empirical results showing that the relevance of a topic to a document is greatly affected by the companion topic’s relevance to the same document, and the extent of the impact differs with respect to different companion topics. We further analyse the phenomenon from classical and quantum-like interference perspectives, and connect the phenomenon to nonreality and contextuality in quantum mechanics. We demonstrate that quantum like model fits in the empirical data, could be potentially used for predicting the relevance when interference exists.

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.