Search results

Taking the discipline construction in colleges and universities as the application background, based on the research on data mining technology and decision support system technology, the data generated by university management information system are effectively utilized. The paper aims to discuss these issues.

Based on the Beijing Key Discipline Information Platform as the data source, the decision tree algorithm of data mining is studied. On the basis of decision tree C4.5, the Bayesian theory is applied to the post-pruning operation of the decision tree.

A decision tree post-pruning algorithm based on the Bayesian theory is studied and put forward in order to simplify the decision tree, which improves the generalization ability of the whole algorithm. Finally, the algorithm is used to build the prediction model of key disciplines. Combined with the decision support system architecture, data warehouse and the data mining algorithm constructed by university discipline, based on J2EE standard enterprise system specification, MVC model is applied. Moreover, a prototype system of decision support system for discipline construction in colleges and universities with browser/server (B/S) structure is completed and implemented.

A decision tree post-pruning algorithm based on the Bayesian theory is studied and put forward in order to simplify the decision tree, which improves the generalization ability of the whole algorithm. Finally, the algorithm is used to build the prediction model of key disciplines. Combined with the decision support system architecture, data warehouse and the data mining algorithm constructed by university discipline, based on J2EE standard enterprise system specification, MVC model is applied. Moreover, a prototype system of decision support system for discipline construction in colleges and universities with B/S structure is completed and implemented.

For the case of many content features, This paper aims to investigate which content features in video and text ads more contribute to accurately predicting the success of crowdfunding by comparing prediction models.

With 1,368 features extracted from 15,195 Kickstarter campaigns in the USA, the authors compare base models such as logistic regression (LR) with tree-based homogeneous ensembles such as eXtreme gradient boosting (XGBoost) and heterogeneous ensembles such as XGBoost + LR.

XGBoost shows higher prediction accuracy than LR (82% vs 69%), in contrast to the findings of a previous relevant study. Regarding important content features, humans (e.g. founders) are more important than visual objects (e.g. products). In both spoken and written language, words related to experience (e.g. eat) or perception (e.g. hear) are more important than cognitive (e.g. causation) words. In addition, a focus on the future is more important than a present or past time orientation. Speech aids (see and compare) to complement visual content are also effective and positive tone matters in speech.

This research makes theoretical contributions by finding more important visuals (human) and language features (experience, perception and future time). Also, in a multimodal context, complementary cues (e.g. speech aids) across different modalities help. Furthermore, the noncontent parts of speech such as positive “tone” or pace of speech are important.

Founders are encouraged to assess and revise the content of their video or text ads as well as their basic campaign features (e.g. goal, duration and reward) before they launch their campaigns. Next, overly complex ensembles may suffer from overfitting problems. In practice, model validation using unseen data is recommended.

Rather than reducing the number of content feature dimensions (Kaminski and Hopp, 2020), by enabling advanced prediction models to accommodate many contents features, prediction accuracy rises substantially.

Environmental performance becomes a key issue for the sustainable development. Recently, incremental information technology is adopted to collect environmental data and improve environmental performance. Previous environmental efficiency measures mainly focus on individual decision-making units (DMUs). Benefited from the information technology, this paper develops a new environmental efficiency measure to explore the implicit alliances among DMUs and applies it to Xiangjiang River.

This study formulates a new data envelopment analysis (DEA) environmental cross-efficiency measure that considers DMUs' alliances. Each DMUs' alliance is formulated by the DMUs who are supervised by the same manager. In cross-efficiency evaluation context, this paper adopts DMUs' alliances rather than individual DMUs to derive the environmental cross-efficiency measure considering undesirable outputs. Furthermore, the Tobit regression is conducted to analyze the influence of exogenous factors about the environmental cross-efficiency.

The findings show that (1) Chenzhou performs the best while Xiangtan performed the worst along Xiangjiang River. (2) The environmental efficiency of cities in Xiangjiang River is generally low. Increasing public budgetary expenditure can improve environmental efficiency of cities. (3) The larger the alliance size, the higher environmental efficiency. (4) The income level is negatively correlated with environmental efficiency, indicating that the economy is at the expense of the environment in Xiangjiang River.

This paper contributes to developing a new environmental DEA cross-efficiency measure considering DMUs' alliance, and combining DEA cross-efficiency and Tobit regression in environmental performance measurement of Xiangjiang River. This paper examines the exogenous factors that have influences on environmental efficiency of Xiangjiang River and derive policy implications to improve the sustainable operation.

Corporate social responsibility has gradually become an essential enterprise responsibility under stakeholders’ expectations. Employee care strategies involve both qualitative and quantitative factors and are receiving special attention with the advent of the information age. In previous studies, a company’s policy of employee care may not fit with the needs of the employees. Consequently, the purpose of this paper is to investigate enterprises’ employee care from the employee’s perspective by adopting a hybrid multiple attribute decision making (MADM) model.

This study is based on 159 interviews with senior employees and/or department managers using a survey questionnaire. This study uses the MADM model to conduct the analysis. First, this research study used Decision-Making Trial and Evaluation Laboratory (DEMATEL) to construct an influential network relations map of the 4 dimensions and 13 criteria of employee care. Second, this study uses DEMATEL-based Analytic Network Process to conduct a weight analysis for each dimension and criterion. Third, this study uses VIKOR to calculate employees’ level of satisfaction as well as the gap from the “aspired level.”

The results of the study revealed the critical factors influencing employee care and proposed a systematic plan to be used as a reference for improvement. The improvement sequence revealed the following order: Equal employment opportunities→Good industrial relations and benefits→Responsibility to train and educate employees→Occupational health and safety. The empirical results showed there was still 35 percent room for improvement in the enterprises’ implementation policy of employee care.

The implementation of employee care has become an important issue for corporations since it helps to sustain and to increase an enterprise’s competitiveness in the business environment. However, the extant literature on employee care comes from enterprises’ perspectives instead of from employees’ perspectives. This research investigates the key factors of employee care and successfully shows MADM to be an effective model for the planning and implementation of corporate social responsibilities’ employee care from the perspective of employees.