Yi-Cheng Chen and Yen-Liang Chen
In this “Info-plosion” era, recommendation systems (or recommenders) play a significant role in finding interesting items in the surge of online digital activity and e-commerce…
Abstract
Purpose
In this “Info-plosion” era, recommendation systems (or recommenders) play a significant role in finding interesting items in the surge of online digital activity and e-commerce. The purpose of this paper is to model users' preference evolution to recommend potential items which users may be interested in.
Design/methodology/approach
A novel recommendation system, namely evolution-learning recommendation (ELR), is developed to precisely predict user interest for making recommendations. Differing from prior related methods, the authors integrate the matrix factorization (MF) and recurrent neural network (RNN) to effectively describe the variation of user preferences over time.
Findings
A novel cumulative factorization technique is proposed to efficiently decompose a rating matrix for discovering latent user preferences. Compared to traditional MF-based methods, the cumulative MF could reduce the utilization of computation resources. Furthermore, the authors depict the significance of long- and short-term effects in the memory cell of RNN for evolution patterns. With the context awareness, a learning model, V-LSTM, is developed to dynamically capture the evolution pattern of user interests. By using a well-trained learning model, the authors predict future user preferences and recommend related items.
Originality/value
Based on the relations among users and items for recommendation, the authors introduce a novel concept, virtual communication, to effectively learn and estimate the correlation among users and items. By incorporating the discovered latent features of users and items in an evolved manner, the proposed ELR model could promote “right” things to “right” users at the “right” time. In addition, several extensive experiments are performed on real datasets and are discussed. Empirical results show that ELR significantly outperforms the prior recommendation models. The proposed ELR exhibits great generalization and robustness in real datasets, including e-commerce, industrial retail and streaming service, with all discussed metrics.
Details
Keywords
Yi-Cheng Chen, Yun-Hao Cheng, Jui-Tang Tseng and Kun-Ju Hsieh
This paper aims to present simulation results of a harmonic drive (HD) with involute flexspline (FS) profiles based on two-dimensional (2-D) finite element analysis (FEA).
Abstract
Purpose
This paper aims to present simulation results of a harmonic drive (HD) with involute flexspline (FS) profiles based on two-dimensional (2-D) finite element analysis (FEA).
Design/methodology/approach
First, the mathematical model of the FS with involute tooth profile was developed using a straight-edge rack cutter based on the theory of gearing. Then the engaging circular spline (CS) with conjugate tooth profile of FS was derived based on the enveloping theory and theory of gearing. Additionally, a mesh generation program was developed to discretize the FS based on the mathematical model. An elliptical wave generator (WG) was inserted into the FS, and a torque was applied to drive the FS meshing with the CS. The WG and the CS were both assumed to be rigid in the finite element model.
Findings
Finally, a 2-D FEA was conducted to explore the stress distribution on the FS, the engagement movement of the FS, the torsional stiffness and the engaged area of teeth of the HD under various conditions. Moreover, this research also studied the effect of changing pressure angle of the involute FS on the performance of the HD.
Research limitations/implications
The simulation model and methodology presented in this paper paved the way for further investigation and optimization of the HD with involute tooth profile FS and conjugate CS.
Originality/value
The simulation model of HD is established on conjugate shape based on the theory of gearing and an automatic mesh generation program is developed to generate the finite element model. The characteristics of the HD can thus be simulated according to the developed model.
Details
Keywords
Tipajin Thaipisutikul and Yi-Cheng Chen
Tourism spot or point-of-interest (POI) recommendation has become a common service in people's daily life. The purpose of this paper is to model users' check-in history in order…
Abstract
Purpose
Tourism spot or point-of-interest (POI) recommendation has become a common service in people's daily life. The purpose of this paper is to model users' check-in history in order to predict a set of locations that a user may soon visit.
Design/methodology/approach
The authors proposed a novel learning-based method, the pattern-based dual learning POI recommendation system as a solution to consider users' interests and the uniformity of popular POI patterns when making recommendations. Differing from traditional long short-term memory (LSTM), a new users’ regularity–POIs’ popularity patterns long short-term memory (UP-LSTM) model was developed to concurrently combine the behaviors of a specific user and common users.
Findings
The authors introduced the concept of dual learning for POI recommendation. Several performance evaluations were conducted on real-life mobility data sets to demonstrate the effectiveness and practicability of POI recommendations. The metrics such as hit rate, precision, recall and F-measure were used to measure the capability of ranking and precise prediction of the proposed model over all baselines. The experimental results indicated that the proposed UP-LSTM model consistently outperformed the state-of-the-art models in all metrics by a large margin.
Originality/value
This study contributes to the existing literature by incorporating a novel pattern–based technique to analyze how the popularity of POIs affects the next move of a particular user. Also, the authors have proposed an effective fusing scheme to boost the prediction performance in the proposed UP-LSTM model. The experimental results and discussions indicate that the combination of the user's regularity and the POIs’ popularity patterns in PDLRec could significantly enhance the performance of POI recommendation.
Details
Keywords
Recently, more and more attention has been put forth on the application and deep learning, due to the widespread practicability of neural network computation. The purpose of this…
Abstract
Purpose
Recently, more and more attention has been put forth on the application and deep learning, due to the widespread practicability of neural network computation. The purpose of this paper is developing an effective algorithm to automatically discover the optimal neural network architecture for several real applications.
Design/methodology/approach
The author proposes a novel algorithm, namely, progressive genetic-based neural architecture search (PG-NAS), as a solution to efficiently find the optimal neural network structure for given data. PG-NAS also employs several operations to effectively shrink the search space to reduce the computation cost and improve the accuracy validation.
Findings
The proposed PG-NAS could be utilized on several tasks for discovering the optimal network structure. The author reduces the demand of manual settings when implementing artificial intelligence (AI) models; hence, PG-NAS requires less human intervention than traditional machine learning. The average and top-1 metrics, such as error, loss and accuracy, are used to measure the discovered neural architectures of the proposed model over all baselines. The experimental results show that, with several real datasets, the proposed PG-NAS model consistently outperforms the state-of-the-art models in all metrics.
Originality/value
Generally, the size and the complexity of the search space for the neural network dominates the performance of computation time and resources. In this study, PG-NAS utilizes genetic operations to effectively generate the compact candidate set, i.e. fewer combinations need to be generated when constructing the candidate set. Moreover, by the proposed selector in PG-NAS, the non-promising network structure could be significantly pruned off. In addition, the accuracy derivation of each combination in the candidate set is also a performance bottleneck. The author develops a predictor network to efficiently estimate the accuracy to avoid the time-consuming derivation. The learning of the prediction process is also adjusted dynamically; this adaptive learning of the predictor could capture the pattern of training data effectively and efficiently. Furthermore, the proposed PG-NAS algorithm is applied on several real datasets to show its practicability and scalability.
Details
Keywords
Chia-Yi Cheng and Shang-Ying Chen
This study aims to investigate hazards in theater venues on the performance day by combining operational risk theory with a service blueprint method.
Abstract
Purpose
This study aims to investigate hazards in theater venues on the performance day by combining operational risk theory with a service blueprint method.
Design/methodology/approach
Interviews and Delphi method are applied to find the hazards, then a survey and ANOVA are followed. The study explores a profile of hazards using data from theater venues in Taiwan and examines whether employee characteristics (i.e. professional tasks, experience and working location) affect risk perception.
Findings
The study suggests a new framework represented by a 5 (types of loss events) × 6 (service systems) matrix to check operational risks. The analyses indicate two types of hazards: risk perception about performance and operations by performers and crew (RPPOPC) and audience behaviors and safety (RPABS). RPPOPC is related to the core show, but not all employees possess high RPPOPC. Seniors have relatively low RPPOPC, and frontend house employees possess insufficient RPABS. Further, front house employees, seniors and those working in municipal cities show relatively high RPPOPC in high-loss situations.
Practical implications
Managers can use the analytic framework to effectively identify operational risks in the core show operations and audience service offerings. They can promote risk perception considering employee differences and loss severity. However, the framework does not discuss the cause-and-effect relationship. Incorporating a large amount of loss experience into a risk information system would help clarify this complex relationship.
Originality/value
This study contributes to hazard mitigation in the performing arts sector, both in the peripheral services for customers and in the core show services.
Details
Keywords
Che-Chih Tsao, Ho-Hsin Chang, Meng-Hao Liu, Ho-Chia Chen, Yun-Tang Hsu, Pei-Ying Lin, Yih-Lin Chou, Ying-Chieh Chao, Yun-Hui Shen, Cheng-Yi Huang, Kai-Chiang Chan and Yi-Hung Chen
The purpose of this paper is to propose and demonstrate a new additive manufacturing approach that breaks the layer-based point scanning limitations to increase fabrication speed…
Abstract
Purpose
The purpose of this paper is to propose and demonstrate a new additive manufacturing approach that breaks the layer-based point scanning limitations to increase fabrication speed, obtain better surface finish, achieve material flexibility and reduce equipment costs.
Design/methodology/approach
The freeform additive manufacturing approach conceptually views a 3D article as an assembly of freeform elements distributed spatially following a flexible 3D assembly structure, which conforms to the surface of the article and physically builds the article by sequentially forming the freeform elements by a vari-directional vari-dimensional capable material deposition mechanism. Vari-directional building along tangential directions of part surface gives surface smoothness. Vari-dimensional deposition maximizes material output to increase build rate wherever allowed and minimizes deposition sizes for resolution whenever needed.
Findings
Process steps based on geometric and data processing considerations were described. Dispensing and forming of basic vari-directional and vari-dimensional freeform elements and basic operations of joining them were developed using thermoplastics. Forming of 3D articles at build rates of 2-5 times the fused deposition modeling (FDM) rate was demonstrated and improvement over ten times was shown to be feasible. FDM compatible operations using 0.7 mm wire depositions from a variable exit-dispensing unit were demonstrated. Preliminary tests of a surface finishing process showed a result of 0.8-1.9 um Ra. Initial results of dispensing wax, tin alloy and steel were also shown.
Originality/value
This is the first time that both vari-directional and vari-dimensional material depositions are combined in a new freeform building method, which has potential impact on the FDM and other additive manufacturing methods.
Details
Keywords
This paper aims to explore the relationship between human resource attributes and the operating performances of accounting firms by sampling data from the 2012-2013 Survey Report…
Abstract
Purpose
This paper aims to explore the relationship between human resource attributes and the operating performances of accounting firms by sampling data from the 2012-2013 Survey Report on Accounting Firms, as compiled by the Financial Supervisory Commission in Taiwan.
Design/methodology/approach
Multiple regression analysis is conducted to measure operating performances with various measurements, such as operating profits and business diversification. The independent variables include male to female ratio, percentage of senior executives, percentage of employees with higher education backgrounds, organizational vitality, human resource diversity, percentage of employees with certified public accountant (CPA) qualifications and human resource costs (HRCs). The control variables are the firm history, market shares and ownership structures since the inception of the firms.
Findings
The empirical results regarding the operating profits model suggest that the higher the male to female ratio, the percentage of employees with higher education backgrounds, organizational vitality, human resource diversity, percentage of employees with CPA qualifications and HRCs, the greater the operating profits. Meanwhile, the findings regarding the business diversification model indicate that the higher the male to female ratio, percentage of senior executives and human resource diversity, the greater the business diversification.
Originality/value
It is intended that the research findings can assist the management of accounting firms to understand the human resource attributes critical to operating performances, which will help to enhance the competitiveness of employees, mitigate the operating risks and improve the operating performances of the firms.
Details
Keywords
Sandrotua Bali, Tsai-Ching Chen, Ming-Chou Liu, Suriya Klangrit and Cheng-Yi Lin
With the increasing number of institutions offering online degree programs, there is a growing need to understand the requirements for interactions and the challenges present in…
Abstract
Purpose
With the increasing number of institutions offering online degree programs, there is a growing need to understand the requirements for interactions and the challenges present in online learning environments. Consequently, this qualitative study aims to explore aspects of nontraditional students, typically defined as older than traditional college age, employed full-time or with family responsibilities.
Design/methodology/approach
This study employs a qualitative approach, conducting in-depth interviews with nine nontraditional students. Grounded in social presence theory, this study analyzed the experiences and viewpoints of nontraditional students in the online learning environment, utilizing thematic analysis.
Findings
Thematic analysis unveiled two major themes: interactions in online learning and challenges in online learning. Four sub-themes emerged from interactions in online learning (interaction with instructors, interaction with peers, content interaction and interface interaction). In addition, three sub-themes emerged from challenges in online learning (timing inflexibility, tools and technological barriers and diverse learning modes). The findings of this suggest that nontraditional students derived benefits from online learning, yet they faced limitations in peer interaction and experienced technological barriers.
Originality/value
This study is based on primary data collected from nontraditional students, offering valuable insights into the needs and challenges they face in higher education while engaged in online learning.
Details
Keywords
Prior studies have extensively explored individual examples of unethical behavior in sales organizations but focused little on repeated violation (RV) of ethical codes…
Abstract
Purpose
Prior studies have extensively explored individual examples of unethical behavior in sales organizations but focused little on repeated violation (RV) of ethical codes, particularly when managers develop salesforces. Based on social learning theory (SLT), the authors propose a multilevel model of RV antecedents and suggest that organizational influence (social cues and modeling) and individual factors (observer characteristics and behavioral outcomes) affect RV, especially with increasing recruitment of salespeople.
Design/methodology/approach
Using data from a leading financial company in Taiwan, the authors analyzed 1,231 records of salespeople’s misbehavior through logistic regression and average marginal effects.
Findings
Modeling in the organization (i.e. peer misconduct), observer characteristics (i.e. experience concerning job tenure and prior violations) and behavioral outcomes (i.e. information concealment violations) were all found to affect the likelihood of RV, and the interactional effect of organizational size was confirmed.
Research limitations/implications
This study contributes to ethical decision-making theory by explaining aspects of RV through SLT. Its multilevel model, integrated with organizational strategy theories, adds an SLT-focused paradigm into unethical behavior research by considering vicarious learning and self-learning, alongside the reciprocal determinism of cognition, behavior, and environment.
Practical implications
Managers should consider socially based patterns of violation when initiating a sales business plan. The chances of RV are increased by unethical models in the organization and offenders’ potential for violations, which is reinforced by social environment.
Originality/value
This study clarified the key drivers of RV decision-making using SLT and identified an effective sales development strategy to maintain an ethically responsible salesforce.
Details
Keywords
Fakhar Kamran, Chen Zu‐Chi, Ji Xiaoda and Yi Cheng
Providing a much easier (direct) approach to calculate the Lie point symmetries of (3 + 1) unsteady Navier‐Stokes equations for viscous, incompressible flow in cylindrical polar…
Abstract
Purpose
Providing a much easier (direct) approach to calculate the Lie point symmetries of (3 + 1) unsteady Navier‐Stokes equations for viscous, incompressible flow in cylindrical polar coordinates.
Design/methodology/approach
Lie group theory, is applied to the equations of motion. Symmetries obtained through a direct approach are then used to reduce (3 + 1) Navier‐Stokes system to a system of ordinary differential equations.
Findings
We observed that the approach applied here to calculate the symmetries of the group is entirely straightforward and involves less calculation as compared to the computer programs such as LIE, Symmgrp.max (MACSYMA) or other symbolic manipulation systems. Further, results obtained here will be practical and useful in comprehending the fluid flow behavior.
Research limitations/implications
We only obtained the exact solution through basic transformations (translation and scaling). The similarity reduction through other subalgebras (finite and infinite dimensions) can be used to explore more facts about the Navier‐Stokes equations.
Originality/value
Direct approach provided in this paper can be utilized to achieve symmetries of other physically important PDEs.