Search results

Divisia component data is used in the training of an Aggregate Feedforward Neural Network (AFFNN), a general-purpose connectionist system designed to assist with data mining activities. The neural network is able to learn the money-price relationship, defined as the relationships between the rate of growth of the money supply and inflation. Learned relationships are expressed in terms of an automatically generated series of human-readable and machine-executable rules, shown to meaningfully and accurately describe inflation in terms of the original values of the Divisia component dataset.

Quantitative methods known as scoring models have been traditionally developed for credit granting decisions using statistical procedures. The purpose of this paper is to develop a non‐parametric credit scoring model for micro enterprises that are not maintaining balance sheets, and without having a track record of performance and other credit‐worthy parameters.

Multilayer perceptron procedure is used to evaluate credit reliability in three classes of risk, i.e. bad risk credit, foreclosed risk credit and good risk credit.

The development of a neural network model for micro enterprises facilitates bankers and financial institutions in credit granting decisions in an automatic manner in the Indian context.

This study applies comprehensive information on parameters of financial package prepared by Indian financial institutions and banks to micro enterprises to design a credit risk model. This model, instead of categorizing borrowers in terms of their “ability to pay”, attempts a solution to the unsolved problem of credit availability to micro enterprises in an Indian context, having no past performance track record.

To evaluate past and recent research on the costs of training human resources in Australia and to compare the merits of different research methods used to measure these costs. The discussion is situated in a general context of low employer contribution to training provision in Australia and acute policy debates on public training provision.

The article presents the aggregate results of two recent quantitative surveys of training costs in Australian organizations. Both surveys adopt an economic definition of the costs and concentrate on firm‐specific skills acquired up until new recruits reach average productivity.

Survey results suggest that the informal costs of training human resources outstrip direct training expenditure and average training costs are much larger than commonly assumed in the policy debate in Australia.

Ideally, the surveys reported upon should be extended to include continuing training costs and a measure of the degree of employer‐provided general training.

Official surveys largely underestimate the cost of employer‐provided training in Australia, contributing to (mistaken?) perceptions of private sector disengagement. Existing measures of the costs should adopt a more comprehensive approach, including the use of economic concepts.

This research stresses, both to HR practitioners and policy makers, the value of measuring opportunity costs in training processes, and contributes to its quantification.

Market response modelling is well covered in the marketing literature. However, much less research has been undertaken in the use of neural networks for market response modelling. Describes experiments to fit neural networks to the consumer goods market. Compares the neural network approach with several other possible models. Focuses on the out‐of‐sample performance of the models. Describes a method for adjusting the neural network architecture which leads to better performance on out‐of‐sample data.

The purpose of this paper is to analyze significant determinants to assess the probability of insureds’ intent to buy (ITB) insurance and willingness to pay (WTP) quantum of dollars for security benefits.

The authors use the Double Hurdle Model (DHM) and Neural Network (NN) architecture to analyze the insureds’ behavior for ITB and WTP. The authors apply these frameworks to all the 503 insureds of a branch of a leading insurer in the United Arab Emirates.

The DHM identified age, loans & liabilities, body mass index, travel outside the UAE, salary and country of origin (Middle Eastern and African) as significant determinants to predict WTP for social security benefits. In addition to these determinants, NN architecture identified insurance replacement, holding multiple citizenship, age of parents, mortgages, country of origin: Americas, length of travel, income of previous year and medical conditions of insured as additional important determinants to predict WTP for social security benefits; thus, NN is found to be superior to DHM due to its lowest RMSE and AIC in the holdout sample and also its flexibility and no assumptions unlike econometric models.

Insureds’ data used from one UAE Branch limit the generalizability of empirical findings.

The study findings will enable the insurers to appropriately design the insurance products that match the insurers’ behavior of ITB and WTP for social security benefits.

The study findings have the potential for insurance institutions to be more flexible in their insurance practices through public–private partnerships.

This is the authors’ original research work.

While postdoctoral research (postdoc) training is a common step toward academic careers in science, technology, engineering and mathematics (STEM) fields, the role of postdoc training in social sciences is less clear. An increasing number of social science PhDs are pursuing postdocs. This paper aims to identify factors associated with participation in postdoc training and examines the relationship between postdoc training and subsequent career outcomes, including attainment of tenure-track faculty positions and early career salaries.

Using data from the National Science Foundation Survey of Earned Doctorates and Survey of Doctorate Recipients, this study applies propensity score matching, regression and decomposition analyses to identify the role of postdoc training on the employment outcomes of PhDs in the social science and STEM fields.

Results from the regression analyses indicate that participation in postdoc training is associated with greater PhD research experience, higher departmental research ranking and departmental job placement norms. When the postdocs and non-postdocs groups are balanced on observable characteristics, postdoc training is associated with a higher likelihood of attaining tenure-track faculty positions 7 to 9 years after PhD completion. The salaries of social science tenure-track faculty with postdoc experience eventually surpass the salaries of non-postdoc PhDs, primarily via placement at institutions that offer relatively higher salaries. This pattern, however, does not apply to STEM PhDs.

This study leverages comprehensive, nationally representative data to investigate the role of postdoc training in the career outcomes of social sciences PhDs, in comparison to STEM PhDs. Research findings suggest that for social sciences PhDs interested in academic careers, postdoc training can contribute to the attainment of tenure-track faculty positions and toward earning relatively higher salaries over time. Research findings provide prospective and current PhDs with information helpful in career planning and decision-making. Academic institutions, administrators, faculty and stakeholders can apply these research findings toward developing programs and interventions to provide doctoral students with career guidance and greater career transparency.

Because of the extensive user coverage of news sites and apps, greater social and commercial value can be realized if users can access their favourite news as easily as possible. However, news has a timeliness factor; there are serious cold start and data sparsity in news recommendation, and news users are more susceptible to recent topical news. Therefore, this study aims to propose a personalized news recommendation approach based on topic model and restricted Boltzmann machine (RBM).

Firstly, the model extracts the news topic information based on the LDA2vec topic model. Then, the implicit behaviour data are analysed and converted into explicit rating data according to the rules. The highest weight is assigned to recent hot news stories. Finally, the topic information and the rating data are regarded as the conditional layer and visual layer of the conditional RBM (CRBM) model, respectively, to implement news recommendations.

The experimental results show that using LDA2vec-based news topic as a conditional layer in the CRBM model provides a higher prediction rating and improves the effectiveness of news recommendations.

This study proposes a personalized news recommendation approach based on an improved CRBM. Topic model is applied to news topic extraction and used as the conditional layer of the CRBM. It not only alleviates the sparseness of rating data to improve the efficient in CRBM but also considers that readers are more susceptible to popular or trending news.

The purpose of the study is to address the problems of low accuracy and missed detection of occluded pedestrians and small target pedestrians when using the YOLOX general object detection algorithm for pedestrian detection. This study proposes a multi-level fine-grained YOLOX pedestrian detection algorithm.

First, to address the problem of the original YOLOX algorithm in obtaining a single perceptual field for the feature map before feature fusion, this study improves the PAFPN structure by adding the ResCoT module to increase the diversity of the perceptual field of the feature map and divides the pedestrian multi-scale features into finer granularity. Second, for the CSPLayer of the PAFPN, a weight gain-based normalization-based attention module (NAM) is proposed to make the model pay more attention to the context information when extracting pedestrian features and highlight the salient features of pedestrians. Finally, the authors experimentally determined the optimal values for the confidence loss function.

The experimental results show that, compared with the original YOLOX algorithm, the AP of the improved algorithm increased by 2.90%, the Recall increased by 3.57%, and F1 increased by 2% on the pedestrian dataset.

The multi-level fine-grained YOLOX pedestrian detection algorithm can effectively improve the detection of occluded pedestrians and small target pedestrians.

The authors introduce a multi-level fine-grained ResCoT module and a weight gain-based NAM attention module.

Are the learning procedures of genetic algorithms (GAs) able to generate optimal architectures for artificial neural networks (ANNs) in high frequency data? In this experimental study, GAs are used to identify the best architecture for ANNs. Additional learning is undertaken by the ANNs to forecast daily excess stock returns. No ANN architectures were able to outperform a random walk, despite the finding of non-linearity in the excess returns. This failure is attributed to the absence of suitable ANN structures and further implies that researchers need to be cautious when making inferences from ANN results that use high frequency data.

This study aims to construct a sentiment series generation method for danmu comments based on deep learning, and explore the features of sentiment series after clustering.

This study consisted of two main parts: danmu comment sentiment series generation and clustering. In the first part, the authors proposed a sentiment classification model based on BERT fine-tuning to quantify danmu comment sentiment polarity. To smooth the sentiment series, they used methods, such as comprehensive weights. In the second part, the shaped-based distance (SBD)-K-shape method was used to cluster the actual collected data.

The filtered sentiment series or curves of the microfilms on the Bilibili website could be divided into four major categories. There is an apparently stable time interval for the first three types of sentiment curves, while the fourth type of sentiment curve shows a clear trend of fluctuation in general. In addition, it was found that “disputed points” or “highlights” are likely to appear at the beginning and the climax of films, resulting in significant changes in the sentiment curves. The clustering results show a significant difference in user participation, with the second type prevailing over others.

Their sentiment classification model based on BERT fine-tuning outperformed the traditional sentiment lexicon method, which provides a reference for using deep learning as well as transfer learning for danmu comment sentiment analysis. The BERT fine-tuning–SBD-K-shape algorithm can weaken the effect of non-regular noise and temporal phase shift of danmu text.