A novel movies recommendation algorithm based on reinforcement learning with DDPG policy

2020 ◽  
Vol 13 (1) ◽  
pp. 67-79
Author(s):  
Qiaoling Zhou
Keyword(s):  
Reinforcement Learning ◽  
Learning Algorithm ◽  
Cold Start ◽  
Sparse Data ◽  
Local Optimum ◽  
Recommendation Algorithm ◽  
Data Set ◽  
Content Type ◽  
Recommendation Algorithms ◽  
The Impact

PurposeEnglish original movies played an important role in English learning and communication. In order to find the required movies for us from a large number of English original movies and reviews, this paper proposed an improved deep reinforcement learning algorithm for the recommendation of movies. In fact, although the conventional movies recommendation algorithms have solved the problem of information overload, they still have their limitations in the case of cold start-up and sparse data.Design/methodology/approachTo solve the aforementioned problems of conventional movies recommendation algorithms, this paper proposed a recommendation algorithm based on the theory of deep reinforcement learning, which uses the deep deterministic policy gradient (DDPG) algorithm to solve the cold starting and sparse data problems and uses Item2vec to transform discrete action space into a continuous one. Meanwhile, a reward function combining with cosine distance and Euclidean distance is proposed to ensure that the neural network does not converge to local optimum prematurely.FindingsIn order to verify the feasibility and validity of the proposed algorithm, the state of the art and the proposed algorithm are compared in indexes of RMSE, recall rate and accuracy based on the MovieLens English original movie data set for the experiments. Experimental results have shown that the proposed algorithm is superior to the conventional algorithm in various indicators.Originality/valueApplying the proposed algorithm to recommend English original movies, DDPG policy produces better recommendation results and alleviates the impact of cold start and sparse data.

scholarly journals Research on task offloading based on deep reinforcement learning in mobile edge environment

2020 ◽  
Vol 309 ◽  
pp. 03026
Author(s):  
Xia Gao ◽  
Fangqin Xu
Keyword(s):  
Reinforcement Learning ◽  
Learning Algorithm ◽  
Edge Computing ◽  
Internet Technology ◽  
Good Effect ◽  
Mobile Edge Computing ◽  
Data Set ◽  
The Impact ◽  
Task Offloading ◽  
Reinforcement Learning Algorithm

With the rapid development of Internet technology and mobile terminals, users’ demand for high-speed networks is increasing. Mobile edge computing proposes a distributed caching approach to deal with the impact of massive data traffic on communication networks, in order to reduce network latency and improve user service quality. In this paper, a deep reinforcement learning algorithm is proposed to solve the task unloading problem of multi-service nodes. The simulation platform iFogSim and data set Google Cluster Trace are used to carry out experiments. The final results show that the task offloading strategy based on DDQN algorithm has a good effect on energy consumption and cost, it has verified the application prospect of deep reinforcement learning algorithm in mobile edge computing.

Increasing revenue and attendance in Canadian Baptist churches

2015 ◽  
Vol 42 (12) ◽  
pp. 1071-1089
Author(s):  
Alan Chan ◽  
Bruce G. Fawcett ◽  
Shu-Kam Lee
Keyword(s):  
Church Health ◽  
Eastern Canada ◽  
Data Set ◽  
Content Type ◽  
Factors Affecting ◽  
Scarce Resources ◽  
Baptist Churches ◽  
And Performance ◽  
The Impact ◽  
Simulation Exercises

Purpose – Church giving and attendance are two important indicators of church health and performance. In the literature, they are usually understood to be simultaneously determined. The purpose of this paper is to estimate if there a sustainable church congregation size using Wintrobe’s (1998) dictatorship model. The authors want to examine the impact of youth and adult ministry as well. Design/methodology/approach – Using the data collected from among Canadian Baptist churches in Eastern Canada, this study investigates the factors affecting the level of the two indicators by the panel-instrumental variable technique. Applying Wintrobe’s (1998) political economy model on dictatorship, the equilibrium level of worship attendance and giving is predicted. Findings – Through various simulation exercises, the actual church congregation sizes is approximately 50 percent of the predicted value, implying inefficiency and misallocation of church resources. The paper concludes with insights on effective ways church leaders can allocate scarce resources to promote growth within churches. Originality/value – The authors are the only researchers getting the permission from the Atlantic Canada Baptist Convention to use their mega data set on church giving and congregation sizes as per the authors’ knowledge. The authors are also applying a theoretical model on dictatorship to religious/not for profits organizations.

Robot obstacle avoidance system using deep reinforcement learning

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Xiaojun Zhu ◽  
Yinghao Liang ◽  
Hanxu Sun ◽  
Xueqian Wang ◽  
Bin Ren
Keyword(s):  
Reinforcement Learning ◽  
Collision Avoidance ◽  
Obstacle Avoidance ◽  
Learning Algorithm ◽  
Optimal Path ◽  
Environmental Parameters ◽  
Working Environment ◽  
Content Type ◽  
Practical Applications ◽  
Human Operators

Purpose Most manufacturing plants choose the easy way of completely separating human operators from robots to prevent accidents, but as a result, it dramatically affects the overall quality and speed that is expected from human–robot collaboration. It is not an easy task to ensure human safety when he/she has entered a robot’s workspace, and the unstructured nature of those working environments makes it even harder. The purpose of this paper is to propose a real-time robot collision avoidance method to alleviate this problem. Design/methodology/approach In this paper, a model is trained to learn the direct control commands from the raw depth images through self-supervised reinforcement learning algorithm. To reduce the effect of sample inefficiency and safety during initial training, a virtual reality platform is used to simulate a natural working environment and generate obstacle avoidance data for training. To ensure a smooth transfer to a real robot, the automatic domain randomization technique is used to generate randomly distributed environmental parameters through the obstacle avoidance simulation of virtual robots in the virtual environment, contributing to better performance in the natural environment. Findings The method has been tested in both simulations with a real UR3 robot for several practical applications. The results of this paper indicate that the proposed approach can effectively make the robot safety-aware and learn how to divert its trajectory to avoid accidents with humans within the workspace. Research limitations/implications The method has been tested in both simulations with a real UR3 robot in several practical applications. The results indicate that the proposed approach can effectively make the robot be aware of safety and learn how to change its trajectory to avoid accidents with persons within the workspace. Originality/value This paper provides a novel collision avoidance framework that allows robots to work alongside human operators in unstructured and complex environments. The method uses end-to-end policy training to directly extract the optimal path from the visual inputs for the scene.

The impact of organizational position level and cultural flow direction on the relationship between cultural intelligence and expatriate cross-border adaptation

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ying Zhang ◽  
Yuran Li ◽  
Mark Frost ◽  
Shiyu Rong ◽  
Rong Jiang ◽  
...  
Keyword(s):  
Cultural Intelligence ◽  
Flow Direction ◽  
Power Distance ◽  
Cultural Adjustment ◽  
Data Set ◽  
Content Type ◽  
Cross Border ◽  
Organizational Position ◽  
The Impact ◽  
The Relationship

PurposeThis paper aims to examine the critical role played by cultural flow in fostering successful expatriate cross-border transitions.Design/methodology/approachThe authors develop and test a model on the interplay among cultural intelligence, organizational position level, cultural flow direction and expatriate adaptation, using a data set of 387 expatriate on cross-border transitions along the Belt & Road area.FindingsThe authors find that both organizational position level and cultural flow moderate the relationship between cultural intelligence and expatriate adaptation, whereby the relationship is contingent on the interaction of organizational position status and assignment directions between high power distance and low power distance host environments.Originality/valuePrevious research has shown that higher levels of cultural intelligence are positively related to better expatriate adaptation. However, there is a lack of research on the effect of position difference and cultural flow on such relationship. Our study is among the first to examine how the interaction between cultural flow and organizational position level influences the cultural intelligence (CI) and cultural adjustment relationship in cross-cultural transitions.

Exploring the impact of intellectual capital on company reputation and performance

2018 ◽  
Vol 19 (5) ◽  
pp. 915-934 ◽  
Author(s):  
Gianluca Ginesti ◽  
Adele Caldarelli ◽  
Annamaria Zampella
Keyword(s):  
Human Capital ◽  
Intellectual Capital ◽  
Corporate Reputation ◽  
Intangible Asset ◽  
Competition Authority ◽  
Data Set ◽  
Content Type ◽  
Capital Efficiency ◽  
And Performance ◽  
The Impact

Purpose The purpose of this paper is to analyse the impact of intellectual capital (IC) on the reputation and performance of Italian companies. Design/methodology/approach The paper exploits a unique data set of 452 non-listed companies that obtained a reputational assessment from the Italian Competition Authority (ICA). To test the hypotheses, this study implemented several regression analyses. Findings Results support the argument that human capital efficiency is a key driver of corporate reputation. Findings also reveal that companies, which obtained reputational rating under ICA scrutiny, show a positive relationship between IC elements and various measures of financial performance. Research limitations/implications The study focuses on a single country; it is not free from the imprecisions of Pulic’s VAIC model. Practical implications This paper recommends companies that are interested to achieve a robust reputation should consider the human capital as a strategic intangible asset. Second, the results suggest that companies with an ICA reputational rating are able to leverage their intangibles to potentiate performance and competitiveness. Originality/value This is the first empirical investigation on the contribution of IC in generating value for corporate reputation. Additionally, the study contributes to the literature on the link between IC and performance by examining a sample of firms not yet explored in prior research.

Capital and credit constraints in the engagement of youth in Ghanaian agriculture

2019 ◽  
Vol 80 (1) ◽  
pp. 22-37 ◽  
Author(s):  
Martinson Ankrah Twumasi ◽  
Yuansheng Jiang ◽  
Monica Owusu Acheampong
Keyword(s):  
Positive Impact ◽  
Credit Constraints ◽  
Rural Youth ◽  
Unique Contribution ◽  
Switching Regression ◽  
Credit Constraint ◽  
Data Set ◽  
Content Type ◽  
Application Procedure ◽  
The Impact

Purpose The purpose of this paper is to determine the factors influencing rural youth farmers’ credit constraints status and the effect of credit constraint on the intensity of participation of these farmers in Ghana. Design/methodology/approach The econometric estimation is based on cross-sectional data collected in 2018 from the Brong Ahafo region in Ghana. The sample data set consists of 450 rural youth farmers. The collected data were analyzed through different econometric techniques, using the endogenous switching regression model (ERSM). Findings The direct elicitation approach employed in this study revealed that out of the 450 farmers, 211 (47 percent) of the respondents were credit constrained compared to 239 (53 percent) of their counterparts who were unconstrained. The ERSM indicated that youth farmers education, age, savings, parents occupation reduced the probability of the rural youth farmer to be credit constrained but cumbersome loan application procedure and loan disbursement time positively affect credit constraint. Moreover, farmers that are credit constrained have lower intensity of participation in agriculture activities than a random farmer from the sample. This suggests that access to credit has a positive impact on the intensity of participation in agriculture activities. Research limitations/implications In this study, only rural youth farmers in a particular region were considered. However, there are youths all over the nation. Therefore, future researchers could consider other youth’s farmers elsewhere in the country. Originality/value Although existing studies have examined rural youth farmers’ participation in agriculture and credit constraint separately, the unique contribution of this paper is the analysis of credit constraint of rural youth farmers as well as the impact of credit constraint on the intensity of participation in agriculture activities.

scholarly journals Credit composition and income inequality in Vietnam: an empirical analysis

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Le Quoc Hoi ◽  
Hương Lan Trần
Keyword(s):  
Income Inequality ◽  
Educational Level ◽  
Generalized Method Of Moments ◽  
Institutional Quality ◽  
Primary Data ◽  
Data Set ◽  
Content Type ◽  
Private Credit ◽  
The Impact

PurposeThis paper aims to examine the credit composition and income inequality reduction in Vietnam. In particular, the authors focus on the distinction between policy and commercial credits and investigate whether these two types of credit had adverse effects on income inequality. The authors also examine whether the impact of policy credit on income inequality is conditioned by the educational level and institutional quality.Design/methodology/approachThe authors use the primary data set, which contains a panel of 60 provinces collected from the General Statistics Office of Vietnam from 2002 to 2016. The authors employ the generalized method of moments to solve the endogenous problem.FindingsThe authors show that while commercial credit increases income inequality, policy credit contributes to reducing income inequality in Vietnam. In addition, we provide evidence that the institutional quality and educational level condition the impact of policy credit on income inequality. Based on the findings, the paper implies that it was not the size of the private credit but its composition that mattered in reducing income inequality, due to the asymmetric effects of different types of credit.Originality/valueThis is the first study that examines the links between the two components of credit and income inequality as well as constraints of the links. The authors argue that analyzing the separate effects of commercial and policy credits is more important for explaining the role of credit in income inequality than the size of total credit.

scholarly journals Revenue diversification and financial sustainability of microfinance institutions

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Peter Nderitu Githaiga
Keyword(s):  
Generalized Method Of Moments ◽  
Financial Sustainability ◽  
Microfinance Institutions ◽  
Revenue Diversification ◽  
Estimation Model ◽  
Data Set ◽  
Content Type ◽  
Method Of Moments Estimation ◽  
The Impact ◽  
Positive Effect

PurposeThis paper aims to investigate whether revenue diversification affects the financial sustainability of microfinance institutions (MFIs).Design/methodology/approachThe study uses a worldwide panel data set of 443 MFIs in 108 countries for the period 2013–2018 and two-step system Generalized Method of Moments estimation model.FindingsThe study finds that revenue diversification has a significant and positive effect on the financial sustainability of MFIs.Practical implicationsThe findings of this study actually offer important managerial and policy lessons on MFIs’ financial sustainability. Microfinance managers and policymakers should consider revenue diversification as a strategy through which MFIs can attain financial sustainability instead of overreliance on donations and government subsidiesOriginality/valueUnlike previous studies that examined revenue diversification in the context of banking firms, this study contributes to literature by examining the impact of revenue diversification of the financial sustainability of MFIs.

scholarly journals Collaborative Filtering Recommendation Algorithm Based on TF-IDF and User Characteristics

2021 ◽  
Vol 11 (20) ◽  
pp. 9554
Author(s):  
Jianjun Ni ◽  
Yu Cai ◽  
Guangyi Tang ◽  
Yingjuan Xie
Keyword(s):  
Collaborative Filtering ◽  
User Characteristics ◽  
Recommendation Algorithm ◽  
Data Set ◽  
Rating Data ◽  
User Similarity ◽  
Public Data ◽  
Weighted Algorithm ◽  
Collaborative Filtering Algorithm ◽  
The Impact

The recommendation algorithm is a very important and challenging issue for a personal recommender system. The collaborative filtering recommendation algorithm is one of the most popular and effective recommendation algorithms. However, the traditional collaborative filtering recommendation algorithm does not fully consider the impact of popular items and user characteristics on the recommendation results. To solve these problems, an improved collaborative filtering algorithm is proposed, which is based on the Term Frequency-Inverse Document Frequency (TF-IDF) method and user characteristics. In the proposed algorithm, an improved TF-IDF method is used to calculate the user similarity on the basis of rating data first. Secondly, the multi-dimensional characteristics information of users is used to calculate the user similarity by a fuzzy membership method. Then, the above two user similarities are fused based on an adaptive weighted algorithm. Finally, some experiments are conducted on the movie public data set, and the experimental results show that the proposed method has better performance than that of the state of the art.

scholarly journals Tag-Aware Recommender System Based on Deep Reinforcement Learning

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Zhiruo Zhao ◽  
Xiliang Chen ◽  
Zhixiong Xu ◽  
Lei Cao
Keyword(s):  
Reinforcement Learning ◽  
Recommender System ◽  
Complex Function ◽  
Cold Start ◽  
Sparse Data ◽  
Excellent Performance ◽  
Minimum Error ◽  
Reward Function ◽  
Function Design

Recently, the application of deep reinforcement learning in the recommender system is flourishing and stands out by overcoming drawbacks of traditional methods and achieving high recommendation quality. The dynamics, long-term returns, and sparse data issues in the recommender system have been effectively solved. But the application of deep reinforcement learning brings problems of interpretability, overfitting, complex reward function design, and user cold start. This study proposed a tag-aware recommender system based on deep reinforcement learning without complex function design, taking advantage of tags to make up for the interpretability problems existing in the recommender system. Our experiment is carried out on the MovieLens dataset. The result shows that the DRL-based recommender system is superior than traditional algorithms in minimum error, and the application of tags have little effect on accuracy when making up for interpretability. In addition, the DRL-based recommender system has excellent performance on user cold start problems.

Sign in / Sign up

Export Citation Format

Share Document