Learning domain-specific sentiment lexicons for predicting product sales

Generic sentiment lexicons have been widely used for sentiment analysis these days. However, manually constructing sentiment lexicons is very time-consuming and it may not be feasible for certain application domains where annotation expertise is not available. One contribution of this paper is the development of a statistical learning based computational method for the automatic construction of domain-specific sentiment lexicons to enhance cross-domain sentiment analysis. Our initial experiments show that the proposed methodology can automatically generate domain-specific sentiment lexicons which contribute to improve the effectiveness of opinion retrieval at the document level. Another contribution of our work is that we show the feasibility of applying the sentiment metric derived based on the automatically constructed sentiment lexicons to predict product sales of certain product categories. Our research contributes to the development of more effective sentiment analysis system to extract business intelligence from numerous opinionated expressions posted to the Web

A proposed validation framework for expert elicited Bayesian Networks

The popularity of Bayesian Network modelling of complex domains using expert elicitation has raised questions of how one might validate such a model given that no objective dataset exists for the model. Past attempts at delineating a set of tests for establishing confidence in an entirely expert-elicited model have focused on single types of validity stemming from individual sources of uncertainty within the model. This paper seeks to extend the frameworks proposed by earlier researchers by drawing upon other disciplines where measuring latent variables is also an issue. We demonstrate that even in cases where no data exist at all there is a broad range of validity tests that can be used to establish confidence in the validity of a Bayesian Belief Network.

Automated support for versioning compliance checking between workflow management and product lifecycle management (in Chinese)

With the development of enterprise informatisation, Product Lifecycle Management (PLM) systems have been widely deployed and applied in enterprises. This paper analyzes the requirement that conducting version operations on business objects as specified in process models should be compliant with the versioning policies imposed by product lifecycles. This leads to the introduction of the concept of versioning compliance, and the approach of compliance checking that we proposed in our earlier work, which comprises both syntactical compatibility and behavioural compatibility checking. The paper then focuses on the tool implementation for providing automated support to the versioning compliance checking. An empirical evaluation of the tool was also performed with industrial partners using the well-known questionnaire-based method. The evaluation and feedback from practitioners further evidence the practical significance of this research question in the PLM field and demonstrate that the proposed solution with its automated tool support possesses a high application potential.

Validation of scales measuring self-efficacy and outcome expectancy in evidence-based practice

Background: Evidence-based practice (EBP) is embraced internationally as an ideal approach to improve patient outcomes and provide cost-effective care. However, despite the support for and apparent benefits of evidence-based practice, it has been shown to be complex and difficult to incorporate into the clinical setting. Research exploring implementation of evidence-based practice has highlighted many internal and external barriers including clinicians’ lack of knowledge and confidence to integrate EBP into their day-to-day work. Nurses in particular often feel ill-equipped with little confidence to find, appraise and implement evidence. Aims: The following study aimed to undertake preliminary testing of the psychometric properties of tools that measure nurses’ self-efficacy and outcome expectancy in regard to evidence-based practice. Methods: A survey design was utilised in which nurses who had either completed an EBP unit or were randomly selected from a major tertiary referral hospital in Brisbane, Australia were sent two newly developed tools: 1) Self-efficacy in Evidence-Based Practice (SE-EBP) scale and 2) Outcome Expectancy for Evidence-Based Practice (OE-EBP) scale. Results: Principal Axis Factoring found three factors with eigenvalues above one for the SE-EBP explaining 73% of the variance and one factor for the OE-EBP scale explaining 82% of the variance. Cronbach’s alpha for SE-EBP, three SE-EBP factors and OE-EBP were all >.91 suggesting some item redundancy. The SE-EBP was able to distinguish between those with no prior exposure to EBP and those who completed an introductory EBP unit. Conclusions: While further investigation of the validity of these tools is needed, preliminary testing indicates that the SE-EBP and OE-EBP scales are valid and reliable instruments for measuring health professionals’ confidence in the process and the outcomes of basing their practice on evidence.

Validation of a new hand-held electronic data capture method for continuous monitoring of subjective appetite sensations

Background When large scale trials are investigating the effects of interventions on appetite, it is paramount to efficiently monitor large amounts of human data. The original hand-held Electronic Appetite Ratings System (EARS) was designed to facilitate the administering and data management of visual analogue scales (VAS) of subjective appetite sensations. The purpose of this study was to validate a novel hand-held method (EARS II (HP® iPAQ)) against the standard Pen and Paper (P&P) method and the previously validated EARS. Methods Twelve participants (5 male, 7 female, aged 18-40) were involved in a fully repeated measures design. Participants were randomly assigned in a crossover design, to either high fat (>48% fat) or low fat (<28% fat) meal days, one week apart and completed ratings using the three data capture methods ordered according to Latin Square. The first set of appetite sensations was completed in a fasted state, immediately before a fixed breakfast. Thereafter, appetite sensations were completed every thirty minutes for 4h. An ad libitum lunch was provided immediately before completing a final set of appetite sensations. Results Repeated measures ANOVAs were conducted for ratings of hunger, fullness and desire to eat. There were no significant differences between P&P compared with either EARS or EARS II (p > 0.05). Correlation coefficients between P&P and EARS II, controlling for age and gender, were performed on Area Under the Curve ratings. R2 for Hunger (0.89), Fullness (0.96) and Desire to Eat (0.95) were statistically significant (p < 0.05). Conclusions EARS II was sensitive to the impact of a meal and recovery of appetite during the postprandial period and is therefore an effective device for monitoring appetite sensations. This study provides evidence and support for further validation of the novel EARS II method for monitoring appetite sensations during large scale studies. The added versatility means that future uses of the system provides the potential to monitor a range of other behavioural and physiological measures often important in clinical and free living trials.

Validation of de-identified record linkage to ascertain hospital admissions in a cohort study

Background Cohort studies can provide valuable evidence of cause and effect relationships but are subject to loss of participants over time, limiting the validity of findings. Computerised record linkage offers a passive and ongoing method of obtaining health outcomes from existing routinely collected data sources. However, the quality of record linkage is reliant upon the availability and accuracy of common identifying variables. We sought to develop and validate a method for linking a cohort study to a state-wide hospital admissions dataset with limited availability of unique identifying variables. Methods A sample of 2000 participants from a cohort study (n = 41 514) was linked to a state-wide hospitalisations dataset in Victoria, Australia using the national health insurance (Medicare) number and demographic data as identifying variables. Availability of the health insurance number was limited in both datasets; therefore linkage was undertaken both with and without use of this number and agreement tested between both algorithms. Sensitivity was calculated for a sub-sample of 101 participants with a hospital admission confirmed by medical record review. Results Of the 2000 study participants, 85% were found to have a record in the hospitalisations dataset when the national health insurance number and sex were used as linkage variables and 92% when demographic details only were used. When agreement between the two methods was tested the disagreement fraction was 9%, mainly due to "false positive" links when demographic details only were used. A final algorithm that used multiple combinations of identifying variables resulted in a match proportion of 87%. Sensitivity of this final linkage was 95%. Conclusions High quality record linkage of cohort data with a hospitalisations dataset that has limited identifiers can be achieved using combinations of a national health insurance number and demographic data as identifying variables.

Preliminary validation of a novel high-resolution melt-based typing method based on the multilocus sequence typing scheme of Streptococcus pyogenes

The major limitation of current typing methods for Streptococcus pyogenes, such as emm sequence typing and T typing, is that these are based on regions subject to considerable selective pressure. Multilocus sequence typing (MLST) is a better indicator of the genetic backbone of a strain but is not widely used due to high costs. The objective of this study was to develop a robust and cost-effective alternative to S. pyogenes MLST. A 10-member single nucleotide polymorphism (SNP) set that provides a Simpson’s Index of Diversity (D) of 0.99 with respect to the S. pyogenes MLST database was derived. A typing format involving high-resolution melting (HRM) analysis of small fragments nucleated by each of the resolution-optimized SNPs was developed. The fragments were 59–119 bp in size and, based on differences in G+C content, were predicted to generate three to six resolvable HRM curves. The combination of curves across each of the 10 fragments can be used to generate a melt type (MelT) for each sequence type (ST). The 525 STs currently in the S. pyogenes MLST database are predicted to resolve into 298 distinct MelTs and the method is calculated to provide a D of 0.996 against the MLST database. The MelTs are concordant with the S. pyogenes population structure. To validate the method we examined clinical isolates of S. pyogenes of 70 STs. Curves were generated as predicted by G+C content discriminating the 70 STs into 65 distinct MelTs.

Measuring literalism and symbolism in advertisements : scale development and validation

A basic element in advertising strategy is the choice of an appeal. Many researchers have studied communication message form and specifically forms of literalism and symbolism, or some variation. The motives for such study are grounded in increasing the effectiveness of commercial communication messages, especially advertising messages. Advertising research studies typically use forms of literalism (e.g. informativeness) or symbolism (e.g. metaphoric, tropes, schemes figures of speech, and rhetorical figures) as independent variables and compare these against one or more of the traditional advertising effectiveness measures as dependent variable(s). The main challenge in assessing the effectiveness of literalism or symbolism in message content is the discreet identification of the construct. However, no standard, empirically-tested measure was located in the literature.

Increasing loyalty to breastfeeding : investigating a product development strategy

This paper demonstrates how social marketing insights were used to influence women’s loyalty to breastfeeding. The paper reports on a social marketing campaign undertaken by the Australian Breastfeeding Association and a government health department, which used a product development strategy in order to increase breastfeeding loyalty. Seeking new approaches to support breastfeeding behaviors is critical and timely, because while initiation rates of breastfeeding are high in developed countries such as the UK, Australia, Canada, and the USA, duration rates are significantly lower. Results indicate that a product focused strategy influences pregnant women’s loyalty to exclusively breastfeeding.

Interactive product browsing and configuration using remote augmented reality sales services

Real-time remote sales assistance is an underdeveloped component of online sales services. Solutions involving web page text chat, telephony and video support prove problematic when seeking to remotely guide customers in their sales processes, especially with configurations of physically complex artefacts. Recently, there has been great interest in the application of virtual worlds and augmented reality to create synthetic environments for remote sales of physical artefacts. However, there is a lack of analysis and development of appropriate software services to support these processes. We extend our previous work with the detailed design of configuration context services to support the management of an interactive sales session using augmented reality. We detail the context and configuration services required, presenting a novel data service streaming configuration information to the vendor for business analytics. We expect that a fully implemented configuration management service, based on our design, will improve the remote sales experience for both customers and vendors alike via analysis of the streamed information.

A novel cyclization product from the polyphosphoric acid catalyzed addition of propanal to limonene

The polyphosphoric acid catalyzed addition of propanal to limonene yielded a novel bicyclic ether 2,2,6-trimethyl-4-ethyl-3-oxabicyclo[3.3.1]non-6-ene (I). The yield of (I) was significantly increased by carrying out the reaction under nitrogen rather than in air.

A framework for life cycle cost estimation of a product family at the early stage of product development

A cost estimation method is required to estimate the life cycle cost of a product family at the early stage of product development in order to evaluate the product family design. There are difficulties with existing cost estimation techniques in estimating the life cycle cost for a product family at the early stage of product development. This paper proposes a framework that combines a knowledge based system and an activity based costing techniques in estimating the life cycle cost of a product family at the early stage of product development. The inputs of the framework are the product family structure and its sub function. The output of the framework is the life cycle cost of a product family that consists of all costs at each product family level and the costs of each product life cycle stage. The proposed framework provides a life cycle cost estimation tool for a product family at the early stage of product development using high level information as its input. The framework makes it possible to estimate the life cycle cost of various product family that use any types of product structure. It provides detailed information related to the activity and resource costs of both parts and products that can assist the designer in analyzing the cost of the product family design. In addition, it can reduce the required amount of information and time to construct the cost estimation system.

An assessment tool for selection of appropriate daylighting solutions for buildings in tropical and subtropical regions : Validation using radiance simulation

A new decision-making tool that will assist designers in the selection of appropriate daylighting solutions for buildings in tropical locations has been previously proposed by the authors. Through an evaluation matrix that prioritizes the parameters that best respond to the needs of tropical climates (e.g. reducing solar gain and protection from glare) the tool determines the most appropriate devices for specific climate and building inputs. The tool is effective in demonstrating the broad benefits and limitations of the different daylight strategies for buildings in the tropics. However for thorough analysis and calibration of the tool, validation is necessary. This paper presents a first step in the validation process. RADIANCE simulations were conducted to compare simulation performance with the performance predicted by the tool. To this end, an office building case study in subtropical Brisbane, Australia, and five different daylighting devices including openings, light guiding systems and light transport systems were simulated. Illuminance, light uniformity, daylight penetration and glare analysis were assessed for each device. The results indicate the tool can appropriately rank and recommend daylighting strategies based on specific building inputs for tropical and subtropical regions, making it a useful resource for designers.