Multimodal semantic-associative collateral labelling and indexing of still images

A novel framework for multimodal semantic-associative collateral image labelling, aiming at associating image regions with textual keywords, is described. Both the primary image and collateral textual modalities are exploited in a cooperative and complementary fashion. The collateral content and context based knowledge is used to bias the mapping from the low-level region-based visual primitives to the high-level visual concepts defined in a visual vocabulary. We introduce the notion of collateral context, which is represented as a co-occurrence matrix, of the visual keywords, A collaborative mapping scheme is devised using statistical methods like Gaussian distribution or Euclidean distance together with collateral content and context-driven inference mechanism. Finally, we use Self Organising Maps to examine the classification and retrieval effectiveness of the proposed high-level image feature vector model which is constructed based on the image labelling results.

Codification vs personalisation: A study of the information evaluation practice between aerospace and construction industries

In the emerging digital economy, the management of information in aerospace and construction organisations is facing a particular challenge due to the ever-increasing volume of information and the extensive use of information and communication technologies (ICTs). This paper addresses the problems of information overload and the value of information in both industries by providing some cross-disciplinary insights. In particular it identifies major issues and challenges in the current information evaluation practice in these two industries. Interviews were conducted to get a spectrum of industrial perspectives (director/strategic, project management and ICT/document management) on these issues in particular to information storage and retrieval strategies and the contrasting approaches to knowledge and information management of personalisation and codification. Industry feedback was collected by a follow-up workshop to strengthen the findings of the research. An information-handling agenda is outlined for the development of a future Information Evaluation Methodology (IEM) which could facilitate the practice of the codification of high-value information in order to support through-life knowledge and information management (K&IM) practice.

Flexible MMSE DFE using length and LAG adaptation

This paper investigates how to choose the optimum tap-length and decision delay for the decision feedback equalizer (DFE). Although the feedback filter length can be set as the channel memory, there is no closed-form expression for the feedforward filter length and decision delay. In this paper, first we analytically show that the two dimensional search for the optimum feedforward filter length and decision delay can be simplified to a one dimensional search, and then describe a new adaptive DFE where the optimum structural parameters can be self-adapted.

Emotion and anticipation in an enactive framework for cognition (response to Andy Clark)

Undeniably, anticipation plays a crucial role in cognition. By what means, to what extent, and what it achieves remain open questions. In a recent BBS target article, Clark (in press) depicts an integrative model of the brain that builds on hierarchical Bayesian models of neural processing (Rao and Ballard, 1999; Friston, 2005; Brown et al., 2011), and their most recent formulation using the free-energy principle borrowed from thermodynamics (Feldman and Friston, 2010; Friston, 2010; Friston et al., 2010). Hierarchical generative models of cognition, such as those described by Clark, presuppose the manipulation of representations and internal models of the world, in as much detail as is perceptually available. Perhaps surprisingly, Clark acknowledges the existence of a “virtual version of the sensory data” (p. 4), but with no reference to some of the historical debates that shaped cognitive science, related to the storage, manipulation, and retrieval of representations in a cognitive system (Shanahan, 1997), or accounting for the emergence of intentionality within such a system (Searle, 1980; Preston and Bishop, 2002). Instead of demonstrating how this Bayesian framework responds to these foundational questions, Clark describes the structure and the functional properties of an action-oriented, multi-level system that is meant to combine perception, learning, and experience (Niedenthal, 2007).

Behavioral and spatial analysis of extraterritorial movements in red foxes (Vulpes vulpes)

Abstract: Movements away from the natal or home territory are important to many ecological processes, including gene flow, population regulation, and disease epidemiology, yet quantitative data on these behaviors are lacking. Red foxes exhibit 2 periods of extraterritorial movements: when an individual disperses and when males search neighboring territories for extrapair copulations during the breeding season. Using radiotracking data collected at 5-min interfix intervals, we compared movement parameters, including distance moved, speed of movement, and turning angles, of dispersal and reproductive movements to those made during normal territorial movements; the instantaneous separation distances of dispersing and extraterritorial movements to the movements of resident adults; and the frequency of locations of 95%, 60%, and 30% harmonic mean isopleths of adult fox home territories to randomly generated fox movements. Foxes making reproductive movements traveled farther than when undertaking other types of movement, and dispersal movements were straighter. Reproductive and dispersal movements were faster than territorial movements and also differed in intensity of search and thoroughness. Foxes making dispersal movements avoided direct contact with territorial adults and moved through peripheral areas of territories. The converse was true for reproductive movements. Although similar in some basic characteristics, dispersal and reproductive movements are fundamentally different both behaviorally and spatially and are likely to have different ultimate purposes and contrasting effects on spatial processes such as disease transmission

The outsourcing strategy of local and multinational firms: A supply base perspective

Firms outsource through connecting to local and global supply bases and making such connections produces costs of search and evaluation, which are a function of transaction characteristics and firm capabilities. We argue that firms outsource more when those costs are low. Hence, domestic subsidiaries of multinational firms, with low cost access to both local and global supply bases, outsource more than either domestic firms or foreign subsidiaries, as confirmed by evidence from a large data panel. We also propose that among foreign subsidiaries, distance from the home country co-determines search and evaluation costs such that subsidiaries from more distant countries outsource less. This is confirmed for geographic distance, but a positive effect is found for political distance and a mixed effect for cultural distance.

Analysing the factors that influence tag choice based on semiotic analysis and activity theory

Social tagging has become very popular around the Internet as well as in research. The main idea behind tagging is to allow users to provide metadata to the web content from their perspective to facilitate categorization and retrieval. There are many factors that influence users' tag choice. Many studies have been conducted to reveal these factors by analysing tagging data. This paper uses two theories to identify these factors, namely the semiotics theory and activity theory. The former treats tags as signs and the latter treats tagging as an activity. The paper uses both theories to analyse tagging behaviour by explaining all aspects of a tagging system, including tags, tagging system components and the tagging activity. The theoretical analysis produced a framework that was used to identify a number of factors. These factors can be considered as categories that can be consulted to redirect user tagging choice in order to support particular tagging behaviour, such as cross-lingual tagging.

Capturing and sharing our collective expertise on climate data: the CHARMe project

For users of climate services, the ability to quickly determine the datasets that best fit one's needs would be invaluable. The volume, variety and complexity of climate data makes this judgment difficult. The ambition of CHARMe ("Characterization of metadata to enable high-quality climate services") is to give a wider interdisciplinary community access to a range of supporting information, such as journal articles, technical reports or feedback on previous applications of the data. The capture and discovery of this "commentary" information, often created by data users rather than data providers, and currently not linked to the data themselves, has not been significantly addressed previously. CHARMe applies the principles of Linked Data and open web standards to associate, record, search and publish user-derived annotations in a way that can be read both by users and automated systems. Tools have been developed within the CHARMe project that enable annotation capability for data delivery systems already in wide use for discovering climate data. In addition, the project has developed advanced tools for exploring data and commentary in innovative ways, including an interactive data explorer and comparator ("CHARMe Maps") and a tool for correlating climate time series with external "significant events" (e.g. instrument failures or large volcanic eruptions) that affect the data quality. Although the project focuses on climate science, the concepts are general and could be applied to other fields. All CHARMe system software is open-source, released under a liberal licence, permitting future projects to re-use the source code as they wish.

Independent uncertainty estimates for coefficient based sea surface temperature retrieval from the Along-Trck Scanning Radiometer instruments

We establish a methodology for calculating uncertainties in sea surface temperature estimates from coefficient based satellite retrievals. The uncertainty estimates are derived independently of in-situ data. This enables validation of both the retrieved SSTs and their uncertainty estimate using in-situ data records. The total uncertainty budget is comprised of a number of components, arising from uncorrelated (eg. noise), locally systematic (eg. atmospheric), large scale systematic and sampling effects (for gridded products). The importance of distinguishing these components arises in propagating uncertainty across spatio-temporal scales. We apply the method to SST data retrieved from the Advanced Along Track Scanning Radiometer (AATSR) and validate the results for two different SST retrieval algorithms, both at a per pixel level and for gridded data. We find good agreement between our estimated uncertainties and validation data. This approach to calculating uncertainties in SST retrievals has a wider application to data from other instruments and retrieval of other geophysical variables.

Sampling uncertainty in gridded sea surface temperature products and Advanced Very High Resolution Radiometer (AVHRR) Global Area Coverage (GAC) data

Sea surface temperature (SST) data are often provided as gridded products, typically at resolutions of order 0.05 degrees from satellite observations to reduce data volume at the request of data users and facilitate comparison against other products or models. Sampling uncertainty is introduced in gridded products where the full surface area of the ocean within a grid cell cannot be fully observed because of cloud cover. In this paper we parameterise uncertainties in SST as a function of the percentage of clear-sky pixels available and the SST variability in that subsample. This parameterisation is developed from Advanced Along Track Scanning Radiometer (AATSR) data, but is applicable to all gridded L3U SST products at resolutions of 0.05-0.1 degrees, irrespective of instrument and retrieval algorithm, provided that instrument noise propagated into the SST is accounted for. We also calculate the sampling uncertainty of ~0.04 K in Global Area Coverage (GAC) Advanced Very High Resolution Radiometer (AVHRR) products, using related methods.

An ontological approach to competency management

Competency management is a very important part of a well-functioning organisation. Unfortunately competency descriptions are not uniformly specified nor defined across borders: National, sectorial or organisational, leading to an opaque competency description market with a multitude of competency frameworks and competency benchmarks. An ontology is a formalised description of a domain, which enables automated reasoning engines to be built which by utilising the interrelations between entities can make “intelligent” choices in different situations within the domain. Introducing formalised competency ontologies automated tools, such as skill gap analysis, training suggestion generation, job search and recruitment, can be developed, which compare and contrast different competency descriptions on the semantic level. The major problem with defining a common formalised ontology for competencies is that there are so many viewpoints of competencies and competency frameworks. Work within the TRACE project has focused on finding common trends within different competency frameworks in order to allow an intermediate competency description to be made, which other frameworks can reference. This research has shown that competencies can be divided up into “knowledge”, “skills” and what we call “others”. An ontology has been created based on this with a simple structure of different “kinds” of “knowledges” and “skills” using semantic interrelations to define the basic semantic structure of the ontology. A prototype tool for analysing a skill gap analysis has been developed. Personal profiles can be produced using the tool and a skill gap analysis is performed on a desired competency profile by using an ontologically based inference engine, which is able to list closest fit and possible proficiency gaps

CAFS in action

For those few readers who do not know, CAFS is a system developed by ICL to search through data at speeds of several million characters per second. Its full name is Content Addressable File Store Information Search Processor, CAFS-ISP or CAFS for short. It is an intelligent hardware-based searching engine, currently available with both ICL's 2966 family of computers and the recently announced Series 39, operating within the VME environment. It uses content addressing techniques to perform fast searches of data or text stored on discs: almost all fields are equally accessible as search keys. Software in the mainframe generates a search task; the CAFS hardware performs the search, and returns the hit records to the mainframe. Because special hardware is used, the searching process is very much more efficient than searching performed by any software method. Various software interfaces are available which allow CAFS to be used in many different situations. CAFS can be used with existing systems without significant change. It can be used to make online enquiries of mainframe files or databases or directly from user written high level language programs. These interfaces are outlined in the body of the report.

Implementing a generic systolic array for genetic algorithms

We have designed a highly parallel design for a simple genetic algorithm using a pipeline of systolic arrays. The systolic design provides high throughput and unidirectional pipelining by exploiting the implicit parallelism in the genetic operators. The design is significant because, unlike other hardware genetic algorithms, it is independent of both the fitness function and the particular chromosome length used in a problem. We have designed and simulated a version of the mutation array using Xilinix FPGA tools to investigate the feasibility of hardware implementation. A simple 5-chromosome mutation array occupies 195 CLBs and is capable of performing more than one million mutations per second. I. Introduction Genetic algorithms (GAs) are established search and optimization techniques which have been applied to a range of engineering and applied problems with considerable success [1]. They operate by maintaining a population of trial solutions encoded, using a suitable encoding scheme.

Systolic random number generation for genetic algorithms

A parallel hardware random number generator for use with a VLSI genetic algorithm processing device is proposed. The design uses an systolic array of mixed congruential random number generators. The generators are constantly reseeded with the outputs of the proceeding generators to avoid significant biasing of the randomness of the array which would result in longer times for the algorithm to converge to a solution. 1 Introduction In recent years there has been a growing interest in developing hardware genetic algorithm devices [1, 2, 3]. A genetic algorithm (GA) is a stochastic search and optimization technique which attempts to capture the power of natural selection by evolving a population of candidate solutions by a process of selection and reproduction [4]. In keeping with the evolutionary analogy, the solutions are called chromosomes with each chromosome containing a number of genes. Chromosomes are commonly simple binary strings, the bits being the genes.

The VERA information environment

We present three components of a virtual research environment developed for the ongoing Roman excavation at Silchester. These components — Recycle Bridge, XDB cross-database search, and Arch3D — provide additional services around the existing core of the system, run on the Integrated Archaeological Database (IADB). They provide, respectively, embedding of legacy applications into portals, cross-database searching, and 3D visualisation of stratigraphic information.