Factors influencing robustness and effectiveness of conditional random fields in active learning frameworks

Active learning approaches reduce the annotation cost required by traditional supervised approaches to reach the same effectiveness by actively selecting informative instances during the learning phase. However, effectiveness and robustness of the learnt models are influenced by a number of factors. In this paper we investigate the factors that affect the effectiveness, more specifically in terms of stability and robustness, of active learning models built using conditional random fields (CRFs) for information extraction applications. Stability, defined as a small variation of performance when small variation of the training data or a small variation of the parameters occur, is a major issue for machine learning models, but even more so in the active learning framework which aims to minimise the amount of training data required. The factors we investigate are a) the choice of incremental vs. standard active learning, b) the feature set used as a representation of the text (i.e., morphological features, syntactic features, or semantic features) and c) Gaussian prior variance as one of the important CRFs parameters. Our empirical findings show that incremental learning and the Gaussian prior variance lead to more stable and robust models across iterations. Our study also demonstrates that orthographical, morphological and contextual features as a group of basic features play an important role in learning effective models across all iterations.

Aspect-based opinion mining from product reviews using conditional random fields

Product reviews are the foremost source of information for customers and manufacturers to help them make appropriate purchasing and production decisions. Natural language data is typically very sparse; the most common words are those that do not carry a lot of semantic content, and occurrences of any particular content-bearing word are rare, while co-occurrences of these words are rarer. Mining product aspects, along with corresponding opinions, is essential for Aspect-Based Opinion Mining (ABOM) as a result of the e-commerce revolution. Therefore, the need for automatic mining of reviews has reached a peak. In this work, we deal with ABOM as sequence labelling problem and propose a supervised extraction method to identify product aspects and corresponding opinions. We use Conditional Random Fields (CRFs) to solve the extraction problem and propose a feature function to enhance accuracy. The proposed method is evaluated using two different datasets. We also evaluate the effectiveness of feature function and the optimisation through multiple experiments.

Evolution of a ~2.7 Ga large igneous province: A volcanological, geochemical and geochronological study of the Agnew Greenstone Belt, and new regional correlations for the Kalgoorlie Terrane (Yilgarn Craton, Western Australia)

The thick package of ~2.7 Ga mafic and ultramafic lavas and intrusions preserved among the Neoarchean of the Kalgoorlie Terrene in Western Australia provides valuable insight into geological processes controlling the most prodigious episode of growth and preservation of juvenile continental crust in Earth’s history. Limited exposure of these rocks results in uncertainty about their age, physical and chemical characteristics, and stratigraphic relationships. This in turn prevents confident correlation of regional occurrences of mafic and ultramafic successions (both intrusive and extrusive) and hinders the interpretation of tectonic setting and magmatic evolution. A recent stratigraphic drilling program of the Neoarchean stratigraphy of the Agnew Greenstone Belt in Western Australia has provided continuous exposures through a c. 7 km thick sequence of mafic and ultramafic units. In this study, we present a volcanological, lithogeochemical and chronological study of the Agnew Greenstone Belt, and provide the first pre-2690 Ma regional correlation across the Kalgoorlie Terrane. The Agnew Greenstone Belt records ~30 m.y. of episodic ultramafic-mafic magmatism that includes two cycles, each defined by a komatiite that is overlain by units that become more evolved and contaminated with time. The sequence is divided into nine conformable packages, each consisting of stacked subaqueous lava flows and comagmatic intrusions, as well as two sills without associated extrusions. Lavas, with the exception of intercalations between two units, form a layer-cake stratigraphy and were likely erupted from a system of fissures tapping the same magma source. The komatiites are not contaminated by continental crust ([La/Sm]PM ~0.7) and are of the Al-undepleted Munro-type. Crustal contamination is evident in many units (Songvang Basalt, Never Can Tell Basalt, Redeemer Basalt, and Turrett Dolerite), as judged by [La/Sm]>1, negative Nb and Ti anomalies, and geochemical mixing trends towards felsic contaminants. Crystal fractionation was also significant, with early olivine and chromite (Mg#>65) followed by plagioclase and clinopyroxene removal (Mg<65), and in the most evolved case, titanomagnetite accumulation. Three new TIMS dates on granophyric zones of mafic sills and one ICP-MS date from an interflow felsic tuff are presented and used for regional stratigraphic correlation. Cycle I magmatism began at ~2720 Ma and ended ~2705 Ma, whereas cycle II began ~2705 Ma and ended at 2690.7±1.2 Ma. Regional correlations indicate the western Kalgoorlie Terrane consists of a remarkably similar stratigraphy that can be recognised at Agnew, Ora Banda and Coolgardie, whereas the eastern part of the terrane (e.g., Kambalda Domain) does not include cycle I, but correlates well with cycle II. This research supports an autochthonous model of greenstone formation, in which one large igneous province, represented by two complete cycles, is constructed on sialic crust. New stratigraphic correlations for the Kalgoorlie Terrane indicate that many units can be traced over distances >100 km, which has implications for exploration targeting for stratigraphically hosted ultramafic Ni and VMS deposits.

Dark-energy constraints and correlations with systematics from CFHTLS weak lensing, SNLS supernovae Ia and WMAP5

Aims We combine measurements of weak gravitational lensing from the CFHTLS-Wide survey, supernovae Ia from CFHT SNLS and CMB anisotropies from WMAP5 to obtain joint constraints on cosmological parameters, in particular, the dark-energy equation-of-state parameter w. We assess the influence of systematics in the data on the results and look for possible correlations with cosmological parameters. Methods We implemented an MCMC algorithm to sample the parameter space of a flat CDM model with a dark-energy component of constant w. Systematics in the data are parametrised and included in the analysis. We determine the influence of photometric calibration of SNIa data on cosmological results by calculating the response of the distance modulus to photometric zero-point variations. The weak lensing data set is tested for anomalous field-to-field variations and a systematic shape measurement bias for high-redshift galaxies. Results Ignoring photometric uncertainties for SNLS biases cosmological parameters by at most 20% of the statistical errors, using supernovae alone; the parameter uncertainties are underestimated by 10%. The weak-lensing field-to-field variance between 1 deg2-MegaCam pointings is 5-15% higher than predicted from N-body simulations. We find no bias in the lensing signal at high redshift, within the framework of a simple model, and marginalising over cosmological parameters. Assuming a systematic underestimation of the lensing signal, the normalisation increases by up to 8%. Combining all three probes we obtain -0.10 < 1 + w < 0.06 at 68% confidence ( -0.18 < 1 + w < 0.12 at 95%), including systematic errors. Our results are therefore consistent with the cosmological constant . Systematics in the data increase the error bars by up to 35%; the best-fit values change by less than 0.15.

Crude oil and agricultural futures: An analysis of correlation dynamics

Correlations between oil and agricultural commodities have varied over previous decades, impacted by renewable fuels policy and turbulent economic conditions. We estimate smooth transition conditional correlation models for 12 agricultural commodities and WTI crude oil. While a structural change in correlations occurred concurrently with the introduction of biofuel policy, oil and food price levels are also key influences. High correlation between biofuel feedstocks and oil is more likely to occur when food and oil price levels are high. Correlation with oil returns is strong for biofuel feedstocks, unlike with other agricultural futures, suggesting limited contagion from energy to food markets.

Deeper and wider fully convolutional network coupled with conditional random fields for scene labeling

Deep convolutional neural networks (DCNNs) have been employed in many computer vision tasks with great success due to their robustness in feature learning. One of the advantages of DCNNs is their representation robustness to object locations, which is useful for object recognition tasks. However, this also discards spatial information, which is useful when dealing with topological information of the image (e.g. scene labeling, face recognition). In this paper, we propose a deeper and wider network architecture to tackle the scene labeling task. The depth is achieved by incorporating predictions from multiple early layers of the DCNN. The width is achieved by combining multiple outputs of the network. We then further refine the parsing task by adopting graphical models (GMs) as a post-processing step to incorporate spatial and contextual information into the network. The new strategy for a deeper, wider convolutional network coupled with graphical models has shown promising results on the PASCAL-Context dataset.

Integrating science through Bayesian Belief Networks : case study of Lyngbya in Moreton Bay

Bayesian Belief Networks (BBNs) are emerging as valuable tools for investigating complex ecological problems. In a BBN, the important variables in a problem are identified and causal relationships are represented graphically. Underpinning this is the probabilistic framework in which variables can take on a finite range of mutually exclusive states. Associated with each variable is a conditional probability table (CPT), showing the probability of a variable attaining each of its possible states conditioned on all possible combinations of it parents. Whilst the variables (nodes) are connected, the CPT attached to each node can be quantified independently. This allows each variable to be populated with the best data available, including expert opinion, simulation results or observed data. It also allows the information to be easily updated as better data become available ----- ----- This paper reports on the process of developing a BBN to better understand the initial rapid growth phase (initiation) of a marine cyanobacterium, Lyngbya majuscula, in Moreton Bay, Queensland. Anecdotal evidence suggests that Lyngbya blooms in this region have increased in severity and extent over the past decade. Lyngbya has been associated with acute dermatitis and a range of other health problems in humans. Blooms have been linked to ecosystem degradation and have also damaged commercial and recreational fisheries. However, the causes of blooms are as yet poorly understood.

Creating a therapeutic environment : a non-randomised controlled trial of a quiet time intervention for patients in acute care

Background: Noise is a significant barrier to sleep for acute care hospital patients, and sleep has been shown to be therapeutic for health, healing and recovery. Scheduled quiet time interventions to promote inpatient rest and sleep have been successfully trialled in critical care but not in acute care settings. Objectives: The study aim was to evaluate as cheduled quiet time intervention in an acute care setting. The study measured the effect of a scheduled quiet time on noise levels, inpatients’ rest and sleep behaviour, and wellbeing. The study also examined the impact of the intervention on patients’, visitors’ and health professionals’ satisfaction, and organisational functioning. Design: The study was a multi-centred non-randomised parallel group trial. Settings: The research was conducted in the acute orthopaedic wards of two major urban public hospitals in Brisbane, Australia. Participants: All patientsadmitted to the two wards in the5-month period of the study were invited to participate, withafinalsample of 299 participants recruited. This sample produced an effect size of 0.89 for an increase in the number of patients asleep during the quiet time. Methods: Demographic data were collected to enable comparison between groups. Data for noise level, sleep status, sleepiness and well being were collected using previously validated instruments: a Castle Model 824 digital sound level indicator; a three point sleep status scale; the Epworth Sleepiness Scale; and the SF12 V2 questionnaire. The staff, patient and visitor surveys on the experimental ward were adapted from published instruments. Results: Significant differences were found between the two groups in mean decibel level and numbers of patients awake and asleep. The difference in mean measured noise levels between the two environments corresponded to a ‘perceived’ difference of 2 to 1. There were significant correlations between average decibel level and number of patients awake and asleep in the experimental group, and between average decibel level and number of patients awake in the control group. Overall, patients, visitors and health professionals were satisfied with the quiet time intervention. Conclusions: The findings show that a quiet time intervention on an acute care hospital ward can affect noise level and patient sleep/wake patterns during the intervention period. The overall strongly positive response from surveys suggests that scheduled quiet time would be a positively perceived intervention with therapeutic benefit.

Measuring Consumer Loyalty : the loyalty orientation scale

Reflecting its importance to thc financial success of organisations, interest in consumer loyalty continues unabated. However, there are still many unanswered questions about its conceptualisation and measurement.These questions must he resolved before academics and practitioners can usefully apply the concept. We argue that consumer loyalty is best conceptualised as a multi-dimensional phenomenon. Based on this multi-dimensional view, we develop and test a new measure of consumer loyalty. We hypothesise a threedimensional structure containing affective, temporal and instrumental dimensions, Results from a preliminary test are reported. The results indicate that the construct can be reprcxeuted with two dimensions: affective and temporal loyally. As an additional check on the reliability of our results, we find significant correlations between these two dimensions and a measure of behavioural loyalty.

Gendered perceptions of experiential value in using web-based retail channels

Purpose – This paper compares the experiential consumption values that motivate consumer choice to purchase online for both male and female purchasers and non-purchasers. Design/methodology/approach – Using the theory of consumption value the study examines gendered perceptions of the functional, social and conditional value of using a virtual consumption setting for purchasing. Data was collected through an online survey and analysed using multiple discriminant analysis to determine meaningful differences between male and female purchasers and non-purchasers. Findings – The findings show that male online purchasers are discriminated from female purchasers by social value and from male non-purchasers by conditional value. Female purchasers are discriminated from male purchasers by functional value and from female non-purchasers by social value. Female non-purchasers are discriminated from female purchasers by conditional value. Male non-purchasers are discriminated from male purchasers by functional and social value. Research limitations/implications – Limitations include using an Internet survey and an Australian sample which may impact the generalisability of the findings to a wider population of Internet users. Future research should involve replication of the study in a country more or less developed in terms of gender composition of internet users to extend the generalisability of the findings. Additionally, researchers should examine whether other dimensions of consumption value,such as social influence through on- and off-line communication networks, may influence consumer choice to purchase online. Practical implications – The study provides practical implications for marketers to leverage consumption values that influence male and female consumers’ choice to purchase online and then drive their behaviour online through integrated marketing campaigns that involve both on- and offline strategies. Originality/value – The research makes an original contribution to the consumer behaviour literature as to date, no research has been found that undertakes such a comprehensive gender-based comparison of the perceived value of using a virtual consumption setting for purchasing.

Combining ganglion cell topology and data of patients with glaucoma to determine a structure-function map

PURPOSE: To introduce techniques for deriving a map that relates visual field locations to optic nerve head (ONH) sectors and to use the techniques to derive a map relating Medmont perimetric data to data from the Heidelberg Retinal Tomograph. METHODS: Spearman correlation coefficients were calculated relating each visual field location (Medmont M700) to rim area and volume measures for 10 degrees ONH sectors (HRT III software) for 57 participants: 34 with glaucoma, 18 with suspected glaucoma, and 5 with ocular hypertension. Correlations were constrained to be anatomically plausible with a computational model of the axon growth of retinal ganglion cells (Algorithm GROW). GROW generated a map relating field locations to sectors of the ONH. The sector with the maximum statistically significant (P < 0.05) correlation coefficient within 40 degrees of the angle predicted by GROW for each location was computed. Before correlation, both functional and structural data were normalized by either normative data or the fellow eye in each participant. RESULTS: The model of axon growth produced a 24-2 map that is qualitatively similar to existing maps derived from empiric data. When GROW was used in conjunction with normative data, 31% of field locations exhibited a statistically significant relationship. This significance increased to 67% (z-test, z = 4.84; P < 0.001) when both field and rim area data were normalized with the fellow eye. CONCLUSIONS: A computational model of axon growth and normalizing data by the fellow eye can assist in constructing an anatomically plausible map connecting visual field data and sectoral ONH data.

The adjustment of leadership styles in intercultural workplace – some evidences from the multinational construction companies in Hong Kong

This study investigates the existence of intercultural adjustment in the multicultural construction workplaces by examining the leadership orientations (task-/people-orientation), communication and conflict resolution skills (high/low-context culture), and power relationship styles (high/low power distance) of local Chinese and the British expatriate project managers in the multinational construction companies in Hong Kong. A sample of project managers (N = 40) and their subordinates (N = 61) were surveyed using the structured questionnaires. Statistical techniques (independent-samples t-test, and Pearson correlation analysis) were employed to evaluate the data. The results revealed a number of interesting findings. First, it was found that both project manager groups equally considered the importance of task performance and interpersonal relationship. The results of correlations analysis provide support for the linkages of the length of working abroad with the change in task/people orientation for Chinese and expatriate managers. The analysis revealed that those Chinese managers who have the longest length of time living or working in Western countries tended to measure higher on task-orientation. Similarly, those British expatriate managers who have the longest period of working in Hong Kong tended to be less task-orientated. Second, local Chinese managers were found to be more confrontational when they strongly disagree with their team members than their British expatriate counterparts. It would appear that stress from project deadline which increase the directness and terseness in communication acts, and retain the composure of project managers in dealing with the subordinates. Finally, our findings show that there is significant difference between local Chinese and British expatriate managers in their power relationship with subordinates. This implies that although the intercultural adjustment might influence perceptions of local and expatriate managers, some dominant deep-rooted cultural values and beliefs are still not easily altered. Conclusions are presented along with suggestions for future studies.

Life cycle modelling and design knowledge development in 3D virtual environments : final report

Experience plays an important role in building management. “How often will this asset need repair?” or “How much time is this repair going to take?” are types of questions that project and facility managers face daily in planning activities. Failure or success in developing good schedules, budgets and other project management tasks depend on the project manager's ability to obtain reliable information to be able to answer these types of questions. Young practitioners tend to rely on information that is based on regional averages and provided by publishing companies. This is in contrast to experienced project managers who tend to rely heavily on personal experience. Another aspect of building management is that many practitioners are seeking to improve available scheduling algorithms, estimating spreadsheets and other project management tools. Such “micro-scale” levels of research are important in providing the required tools for the project manager's tasks. However, even with such tools, low quality input information will produce inaccurate schedules and budgets as output. Thus, it is also important to have a broad approach to research at a more “macro-scale.” Recent trends show that the Architectural, Engineering, Construction (AEC) industry is experiencing explosive growth in its capabilities to generate and collect data. There is a great deal of valuable knowledge that can be obtained from the appropriate use of this data and therefore the need has arisen to analyse this increasing amount of available data. Data Mining can be applied as a powerful tool to extract relevant and useful information from this sea of data. Knowledge Discovery in Databases (KDD) and Data Mining (DM) are tools that allow identification of valid, useful, and previously unknown patterns so large amounts of project data may be analysed. These technologies combine techniques from machine learning, artificial intelligence, pattern recognition, statistics, databases, and visualization to automatically extract concepts, interrelationships, and patterns of interest from large databases. The project involves the development of a prototype tool to support facility managers, building owners and designers. This final report presents the AIMMTM prototype system and documents how and what data mining techniques can be applied, the results of their application and the benefits gained from the system. The AIMMTM system is capable of searching for useful patterns of knowledge and correlations within the existing building maintenance data to support decision making about future maintenance operations. The application of the AIMMTM prototype system on building models and their maintenance data (supplied by industry partners) utilises various data mining algorithms and the maintenance data is analysed using interactive visual tools. The application of the AIMMTM prototype system to help in improving maintenance management and building life cycle includes: (i) data preparation and cleaning, (ii) integrating meaningful domain attributes, (iii) performing extensive data mining experiments in which visual analysis (using stacked histograms), classification and clustering techniques, associative rule mining algorithm such as “Apriori” and (iv) filtering and refining data mining results, including the potential implications of these results for improving maintenance management. Maintenance data of a variety of asset types were selected for demonstration with the aim of discovering meaningful patterns to assist facility managers in strategic planning and provide a knowledge base to help shape future requirements and design briefing. Utilising the prototype system developed here, positive and interesting results regarding patterns and structures of data have been obtained.

Improving the management of building life cycle : a data mining approach

The construction industry has adapted information technology in its processes in terms of computer aided design and drafting, construction documentation and maintenance. The data generated within the construction industry has become increasingly overwhelming. Data mining is a sophisticated data search capability that uses classification algorithms to discover patterns and correlations within a large volume of data. This paper presents the selection and application of data mining techniques on maintenance data of buildings. The results of applying such techniques and potential benefits of utilising their results to identify useful patterns of knowledge and correlations to support decision making of improving the management of building life cycle are presented and discussed.