Conference papers

Exploring Spatial Business Data: a ROA Based eCampus Application

Thanh Thoa Pham Thi et al. — Fri, 22 Feb 2013 07:15:52 PST

In "Smart" environments development, providing users with search utilities for interacting efficiently with web and wireless devices is a key goal. At smaller scales, Google Maps and Google Earth with satellite and street views have helped users for querying general information at specific locations. However, at larger local scales, where detailed 3D geometries linked to business data are needed, there is a recognized lack of related information and functionality for in depth exploration of an area. Linking spatial data and business data helps to enrich the user experience by fulfilling more task specific user needs. This paper presents an eCampus Demonstrator for the National University of Ireland, Maynooth (NUIM) based on a Resource Oriented Architecture (ROA), in which various RESTful web-services have been developed and deployed for querying both spatial data and associated business data. The benefits and drawbacks of the chosen technologies are also discussed. This work can be considered as a platform that can be applied to similar application domains such as exploring business parks, hospitals, museums, etc.

A Service Oriented Architecture for Spatial and Business Data Applications: an eCampus Case Study

James Carswell et al. — Tue, 12 Feb 2013 07:46:21 PST

Location-dependent queries of geospatial data are indispensable for users in today's era of "Smart" environments. At general scales, Google Maps and Google Earth with satellite and street views have provided some utilities for users to query specific locations. However, at local scales, where detailed 3D geometries and linked business data are needed, there is a general lack of related information for in depth exploration of an area. For instance, the following kinds of questions cannot be answered when interacting with Google Maps/Earth on a typical university campus: what classes are scheduled in that room over there? Whose office window is that up there? What buildings/objects can I actually see around me while standing at a specific location on campus? Or in a hospital, by spatial querying hospital datasets we can ask what the surgery schedule is in that operating room? What medical equipment is installed in that room? How far is it to move from this room to another?

Linking spatial data and business data helps to enrich the user experience by fulfilling more user needs, in particular for applications that need to explore detailed local scale data such as in a Zoo, Museum, Hospital, Campus, Business Park, etc. There are normally two types of users for such applications: public users (e.g. visitors to an area) and local users (e.g. staff and students). Access privileges and query levels depend on the user type.

Dynamic Estimation of Rater Reliability in Subjective Tasks Using Multi-Armed Bandits

Alexey Tarasov et al. — Fri, 28 Sep 2012 04:45:55 PDT

Many application areas that use supervised machine learning make use of multiple raters to collect target ratings for training data. Usage of multiple raters, however, inevitably introduces the risk that a proportion of them will be unreliable. The presence of unreliable raters can prolong the rating process, make it more expensive and lead to inaccurate ratings. The dominant, "static" approach of solving this problem in state-of-the-art research is to estimate the rater reliability and to calculate the target ratings when all ratings have been gathered. However, doing it dynamically while raters rate training data can make the acquisition of ratings faster and cheaper compared to static techniques. We propose to cast the problem of the dynamic estimation of rater reliability as a multi-armed bandit problem. Experiments show that the usage of multi-armed bandits for this problem is worthwhile, providing that each rater can rate any asset when asked. The purpose of this paper is to outline the directions of future research in this area.

Dynamic Estimation of Rater Reliability in Regression Tasks using Multi-Armed Bandit Techniques

Alexey Tarasov et al. — Fri, 28 Sep 2012 04:41:08 PDT

A Crowdsourcing Approach to Labelling a Mood Induced Speech Corpus

John Snel et al. — Mon, 28 May 2012 09:30:52 PDT

This paper demonstrates the use of crowdsourcing to accumulate ratings from na ̈ıve listeners as a means to provide labels for a naturalistic emotional speech dataset. In order to do so, listening tasks are performed with a rating tool, which is delivered via the web. The rating requirements are based on the classical dimensions, activation and evaluation, presented to the participant as two discretised 5-point scales. Great emphasis is placed on the participant’s overall understanding of the task, and on the ease-of-use of the tool so that labelling accuracy is reinforced. The accumulation process is ongoing with a goal to supply the research community with a publicly available speech corpus.

Mobile Spatial Interaction in the Future Internet of Things

James Carswell et al. — Thu, 08 Mar 2012 06:17:29 PST

Research and development of mobile information systems in the Future Internet of Things is about delivering technologies built around management and access to real-time heterogeneous datasets. Analyzing these enormous volumes of disparate data on mobile devices requires context-aware smart applications and services. 3DQ (Three Dimensional Query) is our novel mobile spatial interaction (MSI) prototype for data mining and analysis on today’s location and orientation aware “smartphones” within such 3D sensor web environments. Our application tailors a military style threat dome query calculation using MSI with “hidden query removal” functionality to reduce information overload and heighten situation awareness on these commercial off-the-shelf (COTS) devices. Allied MSI research into the information overload problem is ongoing, where map personalisation and other semantic based filtering mechanisms are essential to de-clutter and adapt the exploration of the real world to the processing/display limitations of mobile devices. We propose that another way to filter this information is to intelligently refine the search space. The combined effect gives a more accurate and expected query (search) result for Location-Based Services (LBS) applications by returning information on only those objects/sensor enabled “things” visible within a user’s 3D field-of-view (FOV) as they move through a built environment.

Effects of Variations in 3D Spatial Search Techniques on Mobile Query Speed vs Accuracy

Junjun Yin et al. — Mon, 20 Feb 2012 07:24:19 PST

This paper presents three Spatial Search Algorithms for determiningthe three dimensional visibility shape (threat dome) at a user’s current locationin a built environment. Users then utilize this 3D shape as their query“window” to retrieve information on only those objects visible within a spatialdatabase. Visibility shape searching addresses the information overloadproblem by providing “Hidden Query Removal” functionality for mobile LBS.This functionality will be especially useful in the Web 4.0 era when trillions ofmicro-sensors become available for query through standard IP access.

Asynchronous Ultrasonic Trilateration for Indoor Positioning of Mobile Phones

Viacheslav Filonenko et al. — Mon, 20 Feb 2012 07:24:17 PST

In this paper we discuss how the innate ability of mobile phone speakers to produce ultrasound can be used for accurate indoor positioning. The frequencies in question are in a range between 20 and 22 KHz, which is high enough to be inaudible by humans but still low enough to be generated by today’s mobile phone sound hardware. Our tests indicate that it is possible to generate the given range of frequencies without significant distortions, provided the signal volume is not turned excessively high. In this paper we present and evaluate the accuracy of our asynchronous trilateration method (Lok8) for mobile positioning without requiring knowledge of the time the ultrasonic signal was sent. This approach shows that only the differences in time of arrival to multiple microphones (control points) placed throughout the indoor environment is sufficient. Consequently, any timing issues with client and server synchronization are avoided.

MobiSpatial: Open Source for Mobile Spatial Interaction

Junjun Yin et al. — Wed, 16 Nov 2011 01:53:43 PST

This paper describes our Mobile Spatial Interaction (MSI) prototype MobiSpatial, which benefits from location and orientation aware smartphones and existing open source spatial data initiatives to facilitate user interaction with the geospatial query process. We utilize today’s ubiquitous mobile device as the central computing platform to calculate a mobile user’s visibility shape at his/her current location. MobiSpatial uses this shape as a query “window” in a spatial database to perform line-of-sight, field-of-view and 360º Isovist visibility searches. These visibility based spatial queries reduce the risk of “information overload” by exploiting “hidden query removal” functionality to retrieve only those objects that a user can actually see. By incorporating open source datasets and databases, our application manages to store, index, query, retrieve, and display spatial information solely on the mobile device itself, without concern for any server side computations or internet connections.

Emotional Speech Corpus Construction, Annotation and Distribution

Brian Vaughan et al. — Wed, 05 Oct 2011 02:34:00 PDT

This paper details a process of creating an emotional speech corpus by collecting natural emotional speech assets, analysisng and tagging them (for certain acoustic and linguistic features) and annotating them within an on-line database. The definition of specific metadata for use with an emotional speech corpus is crucial, in that poorly (or inaccurately) annotated assets are of little use in analysis. This problem is compounded by the lack of standardisation for speech corpora, particularly in relation to emotion content. The ISLE Metadata Initiative (IMDI) is the only cohesive attempt at corpus metadata standardisation performed thus far. Although not a comprehensive (or universally adopted) standard, IMDI represents the only current standard for speech corpus metadata available. The adoption of the IMDI standard allows the corpus to be re-used and expanded, in a clear and structured manner, ensuring its re-usability and usefulness as well as addressing issues of data-sparsitiy within the field of emotional speech research.

DITCall-Slow: Slowing Native Speech for Language Learners

Dermot Campbell et al. — Wed, 05 Oct 2011 02:33:58 PDT

It is a common experience of many learners of a foreign language that native speakers (NSs) of that language speak too quickly for them to understand or imitate. Slowing down a segment of speech with older technology results in the familiar deepening of the voice as the pitch drops as well. The result is unpleasant and not particularly instructive. The DITCall-Slow tool slows recorded speech without tonal distortion, so that the learner has – literally – more time to hear what was said by the NS and, especially at slower playback speeds, can attend to the manner in which the sequence was spoken. While the tool is currently being commercialised for the study of English as a foreign language (EFL), it can be applied to any spoken language. The current paper examines why the technology works and its place in modern language learning

Generation of High Quality Audio Natural Emotional Speech Corpus using Task Based Mood Induction

Charlie Cullen et al. — Wed, 05 Oct 2011 02:33:55 PDT

Detecting emotional dimensions [1] in speech is an area of great research interest, notably as a means of improving human computer interaction in areas such as speech synthesis [2]. In this paper, a method of obtaining high quality emotional audio speech assets is proposed. The methods of obtaining emotional content are subject to considerable debate, with distinctions between acted [3] and natural [4] speech being made based on the grounds of authenticity. Mood Induction Procedures (MIP’s) [5] are often employed to stimulate emotional dimensions in a controlled environment. This paper details experimental procedures based around MIP 4, using performance related tasks to engender activation and evaluation responses from the participant. Tasks are specified involving two participants, who must co-operate in order to complete a given task [6] within the allotted time. Experiments designed in this manner also allow for the specification of high quality audio assets (notably 24bit/192Khz [7]), within an acoustically controlled environment [8], thus providing means of reducing unwanted acoustic factors within the recorded speech signal. Once suitable assets are obtained, they will be assessed for the purposes of segregation into differing emotional dimensions. The most statistically robust method of evaluation involves the use of listening tests to determine the perceived emotional dimensions within an audio clip. In this experiment, the FeelTrace [9] rating tool is employed within user listening tests to specify the categories of emotional dimensions for each audio clip.

MEMS: Mobile Environmental Management System

Keith Gardiner et al. — Tue, 04 Oct 2011 07:39:12 PDT

This paper describes a web-based and Mobile Environmental Management System (MEMS) prototype specifically tailored to perform context-aware queries and updating of spatial datasets. Spatially enabled computing can provide situation aware assistance to both web-based and mobile users by presenting the right information at the right time, place, and situation using context-associated knowledge. Contextassociated knowledge is assembled by combining knowledge gained about information accessed in the past with the activities planned by the user, together with other situation dependencies (e.g. location) of these activities. The MEMS datasets are provided by the Canadian Department of Fisheries and Oceans (DFO) and the prototype is customised to the specific needs of the Great Lakes Laboratory for Fisheries and Aquatic Sciences (GLLFAS) Fish Habitat Management Group's requirements for fish species at risk assessment.

SAMATS: Edge Highlighting and Intersection Rating Explained

Joe Hegarty et al. — Tue, 04 Oct 2011 07:39:10 PDT

The creation of detailed 3D buildings models, and to a greater extent the creation of entire city models, has become an area of considerable research over the last couple of decades. The accurate modeling of buildings has LBS (Location Based Services) applications in entertainment, planning, tourism and e-commerce to name just a few. Many modeling systems created to date require manual correspondences to be made across the image set in order to determine the models 3D structure. This paper describes SAMATS, a Semi-Automated Modeling And Texturing System, which has the capability of producing geometrically accurate and photorealistic building models without the need for manual correspondences by using a set of geo-referenced terrestrial images. This paper gives an overview of SAMATS’ components, while describing the Edge Highlighting component and the Intersection Rating step from the Edge Recovery component in detail.

Triangle Grouping and Structure Recovery for 3D Building Modelling and Visualization

Joe Hegarty et al. — Tue, 04 Oct 2011 07:39:08 PDT

Location based and spatial technologies research for the web has endless application for mobile/position content delivery (m-commerce or pcommerce). By exploiting the inherent location-based intelligence of the underling spatial component, relevant examples can include geometrically accurate and photo realistic virtual representations for: property assessments; land/marine information systems; routing information; on-line shopping; cultural heritage/tourist information/sites; etc. A major challenge for this technology is its reliance on professional developers when creating the virtual worlds used for web-based navigation of these services. This paper describes SAMATS, a Semi-Automated Modeling And Texturing System, which has the capability of producing geometrically accurate and photorealistic VR building models for web-based p-commerce applications from a set of geo-referenced terrestrial images. This paper describes the second of three main components that comprise the full functionality of the complete SAMATS implementation. It focuses on the triangle grouping and structure recovery steps, while providing an overview of SAMATS’ other components.

MEMS Mobile GIS: a Spatially Enabled Fish Habitat Management System

Andrea Rizzini et al. — Tue, 04 Oct 2011 07:39:05 PDT

Spatially enabled computing can provide assistance to both web-based and mobile users by exploiting positional information and associated contextual knowledge. The Mobile Environmental Management System (MEMS) is a proof of concept prototype that has been developed in order to simplify administrative duties of biologists at the Department of Fisheries and Oceans (DFO), Canada. MEMS aims to deliver contextaware functionality aided by visualization, analysis and manipulation of spatial and attribute datasets. The resulting application delivers a set of functions and services that aids the DFO’s biologists in making everyday management decisions

An open Approach to Contextualising Heterogeneous Cultural Heritage Datasets

John McAuley et al. — Tue, 04 Oct 2011 07:39:03 PDT

This paper describes a semantic museum application, which aims to present a holistic impression of the Etruscan civilisation. Through the use of a distributed computing paradigm and the CIDOC CRM ontology, the system presents a unified view of a fragmented heritage, while supporting browse and search at a semantic level. Within the cultural heritage world, however, much value is placed on ‘context’, both in describing and presenting heritage artefacts. From this perspective, a platform built upon the distributed search paradigm, although useful in many respects, does not convey how an artefact sits within a broader setting. Narrative concepts are proposed as a way of reconciling heritage artefacts with their original context. A community of domain experts (i.e. Etruscan archaeologists and heritage professionals) is supported in contributing their knowledge and interpretation through a comprehensive authoring process. Narrative content is then organised according to several broad, hierarchically structured topics known as the ‘Sphere of Knowledge’ and a domain ontology describing the artefacts and monument of the Etruscan people. Each artefact is consequently represented through the text and associated with broader topics from the ‘Sphere of Knowledge’. The artefact is therefore not presented in isolation or with lists of similar artefacts but rather discussed from a broader perspective. In our T.Arc.H.N.A system (Towards Archaeological Heritage New Accessibility), annotated narrative content, buttressed by references to real world artefacts, is disseminated to variety of platforms through a semantic web service. The entire approach is developed upon a multi-tiered architecture, allowing for the separation of functionality, yet supporting an open approach to interoperability.

Urban Location Based Services using Mobile Clients: the ICiNG Approach

Eoin Kilfeather et al. — Tue, 04 Oct 2011 07:39:00 PDT

The ICING project is conducting research into eGovernment and Location Based Services and also into two-way interaction with the physical environment. The research focuses on the areas of embedded intelligence, tighter integration of operator platforms and city infrastructure to enable novel services, empowerment of citizens to evolve systems of interaction with the city via social software, input from citizens and sensors for management systems and decision modelling, and a combination of city systems and multi-modal, multi-device communications to provide enhanced services. The technology platforms are gathering indicators from the City, processing the information, proposing actions to be taken with human intervention and supervision and connecting the City with its constituency. Services and information are delivered on a range of commodity devices, providing greater reach and accessibility to local government and communities.

An Open Source Approach to Wireless Positioning Techniques

Seamus Rooney et al. — Tue, 04 Oct 2011 07:38:58 PDT

There are several problems encountered when trying to determine the location of a mobile phone, including weather you are in an urban or rural environment. Also, it is well known that different positioning technologies can work better than others depending on the environment they are in. For example, GPS works well in rural areas but not as well in urban areas, GSM positioning accuracy can be acceptable in urban areas with the right triangulation technology, but is less accurate in rural areas. Positioning with other technologies such as WiFi, Bluetooth, and Semacode all have their own advantages and disadvantages also, depending on the overall environment in which they are used. One research task of the ICiNG project is to address these issues and introduce the next logical step for freely available mobile positioning, advancing the pioneering work done by Place Lab at Intel. The EU-FP6 ICiNG project component that initiates this advance is called the ILC (ICiNG Location Client). The ILC integrates all the above location finding technologies into one positioning module. This paper outlines the technique we developed to combine these technologies and the architecture used to deploy them on a mobile phone. With all these technologies finally available on one device, it is now possible to employ a personal positioning system that can work effectively in any environment. Another important advantage of the ILC is its ability to do this without any direct communication with outside sources, so users need not worry about “big brother” tracking their every movement. The ILC only “listens” for, and makes use of, radio signals that are freely available in the current environment, and does not actively connect to any external network or other services to triangulate its position.

Wireless Positioning Techniques: a Developer's Update

Seamus Rooney et al. — Tue, 04 Oct 2011 07:38:56 PDT

This paper describes the current efforts to develop an open source, privacy sensitive, location determination software component for mobile devices. Currently in mobile computing, the ability of a mobile device to determine its own location is becoming increasingly desirable as the usefulness of such a feature enhances many commercial applications. There have been numerous attempts to achieve this from both the network positioning perspective and also from the wireless beacon angle not to mention the integration of GPS into mobile devices. There are two important aspects to consider when using such a system which are privacy and cost. This paper describes the development of a software component that is sensitive to these issues. The ICiNG Location Client (ILC) is based on some pioneering work carried out by the Place Lab Project at Intel. (Hightower et al., 2006) The ILC advances this research to make it available on mobile devices and attempts to integrate GSM, WiFi, Bluetooth and GPS positioning into one positioning module. An outline of the ILC’s design is given and some of the obstacles encountered during its development are described.