Decision Support for Smart Cities

Permanent URI for this collection


Recent Submissions

Now showing 1 - 10 of 12
  • Item
    Analysis on Evolution Model of Zombie Company under the Absence of Bank Data
    ( 2019-01-08) Fu, Xiangling ; Qi, Jiayin ; Zou, Lulian ; Hou, Yi
    In view of the fact that the data of bank loaning is difficult to be collected, this paper innovatively explores the evolution model of zombie companies by text analysis based on the researches of previous papers at home and abroad based on the financial data of zombie companies. Through the relevant researches on 27 zombies collected, the common characteristics of zombies are found out by the grounded theory. According to the relevant models of enterprise life cycle theory, the evolution model of zombie companies is drawn up, and the corresponding feedback loop of system dynamics causality in each link is further found out, so as to explore the evolutionary rules and the reasons of zombie companies, which is helpful for government to further research on zombies, and favorable for the efficient allocation of market resources and the further rapid development of social economy.
  • Item
    Purchase Prediction Based on a Non-parametric Bayesian Method
    ( 2019-01-08) Liu, Yezheng ; Zhu, Tingting ; Jiang, Yuanchun
    Predicting customer’s next purchase is of paramount importance for online retailers. In this paper, we present a new purchase prediction method to predict customer behavior based on non-parametric Bayesian framework. The proposed method is inspired by topic modeling for text mining. Unlike the conventional methods, we regard customer’s purchase as the result of motivations and automatically determine the number of user purchase motivations. Given customer’s purchase history, we show that customer’s next purchase can be predicted by non-parametric Bayesian model. We apply the model to real-world dataset from and prove it outperforms the traditional methods. Besides that, the proposed method can also determine the number of the motivations owned by users automatically, rendering it a promising approach with a good scalability.
  • Item
    Empirical Research on the Impact of Personalized Recommendation Diversity
    ( 2019-01-08) Zhang, Lin ; Yan, Qiang ; Lu, Junqiang ; Chen, Yongqiang ; Liu, Yi
    Personalized recommendation has important implications in raising online shopping efficiency and increasing product sales. There has been wide interest in finding ways to provide more efficient personalized recommendations. Most existing studies focus on how to improve the accuracy of the recommendation algorithms, or are more concerned on ways to increase consumer satisfaction. Unlike these studies, our study focuses on the process of decision-making, using long tail theory as a basis, to reveal the mechanisms involved in consumers’ adoption of recommendations. This paper analyzes the effect of personalized recommendations from two angles: product sales and ratings, and tries to point out differences in consumer preferences between mainstream products and niche products, high rating products and low rating products, search products and experience products. The study verifies that consumers demand diversity in the recommended content, and also provides suggestions on how to better plan and operate a personalized recommendation system.
  • Item
    A Deep Learning Based Model for Driving Risk Assessment
    ( 2019-01-08) Bian, Yiyang ; Lee, Chang Heon ; Zhao, J Leon ; Wan, Yibo
    In this paper a novel multilayer model is proposed for assessing driving risk. Studying aggressive behavior via massive driving data is essential for protecting road traffic safety and reducing losses of human life and property in smart city context. In particular, identifying aggressive behavior and driving risk are multi-factors combined evaluation process, which must be processed with time and environment. For instance, improper time and environment may facilitate abnormal driving behavior. The proposed Dynamic Multilayer Model consists of identifying instant aggressive driving behavior that can be visited within specific time windows and calculating individual driving risk via Deep Neural Networks based classification algorithms. Validation results show that the proposed methods are particularly effective for identifying driving aggressiveness and risk level via real dataset of 2129 drivers’ driving behavior.
  • Item
    Multi-Source-Data-Oriented Ensemble Learning Based PM 2.5 Concentration Prediction in Shenyang
    ( 2019-01-08) Qi, Tianfang ; Jiang, Hongxun ; Shi, Xiaowen
    Shenyang where is surrounded by smokestack industries and depends on coal heating in winter, is a classical one of cities in China northeastern which has suffered from serious air pollution, especially PM2.5. The existing research on machine learning, based on historical air-monitoring data and meteorological data, does neither forecast accurately nor identify key pollutants for PM2.5. This paper presents a multi-source-data-oriented ensemble learning for predicting PM2.5 concentration. The proposed framework incorporates not only air quality data and weather data, but also industrial emission data, especially those of winter heating enterprises, in Shenyang and nearby cities; the model also takes into account location and emission frequency of pollution sources. All these data are entered into an ensemble learning model based on Extreme Gradient Boosting (XGBoost) in order to predict PM2.5 concentration, which not only improves prediction accuracy effectively, but also provides contribution analysis of different pollutants. Experimental results show that the top two factors affecting PM2.5 concentration are: (1) air pollutant emission quantities and (2) distance from pollution sources to air-monitoring stations. According to the importance of these two factors, we refine feature selection and re-train the ensemble learning model and find that the new model performs better on 72% of evaluation indexes.
  • Item
    Smart Devices and Services for Smart City
    ( 2019-01-08) Ferro, Matteo ; Di Leo, Giuseppe ; Liguori, Consolatina ; Paciello, Vincenzo ; Pietrosanto, Antonio
    Citizen quality of life can be improved through facilities and services that have been thought to ease citizen interaction with municipal authorities, offices and structures. All technologies and devices, used for developing these facilities, are the pillars of the Smart City idea: a City that adapts itself, at least in part, to citizens’ needs. Advanced Metering Infrastructure (AMI) could become the backbone of all the smart city projects. Other public services can be loaded on AMI’s to be smart and thus helping to find the affordability of investments. The paper deals with this topic by describing devices and results of a pilot project, which has been carried out in an Italian middle city (Salerno), to experience the use of RF 169MHz wM-bus based AMI. Experimental results regarding a set of about 2500 installed devices for gas and water metering, car parking management and elder tele-assistance, will be reported in detail to show convenience and problems of this approach.
  • Item
    Spatial-temporal prediction of air quality based on recurrent neural networks
    ( 2019-01-08) Sun, Xiaotong ; Xu, Wei ; Jiang, Hongxun
    To predict air quality (PM2.5 concentrations, et al), many parametric regression models have been developed, while deep learning algorithms are used less often. And few of them takes the air pollution emission or spatial information into consideration or predict them in hour scale. In this paper, we proposed a spatial-temporal GRU-based prediction framework incorporating ground pollution monitoring (GPM), factory emissions (FE), surface meteorology monitoring (SMM) variables to predict hourly PM2.5 concentrations. The dataset for empirical experiments was built based on air quality monitoring in Shenyang, China. Experimental results indicate that our method enables more accurate predictions than all baseline models and by applying the convolutional processing to the GPM and FE variables notable improvement can be achieved in prediction accuracy.
  • Item
    Facilitating Technology Transfer by Patent Knowledge Graph
    ( 2019-01-08) Deng, Weiwei ; Huang, Xiaoming ; Zhu, Peihu
    Technologies are one of the most important driving forces of our societal development and realizing the value of technologies heavily depends on the transfer of technologies. Given the importance of technologies and technology transfer, an increasingly large amount of money has been invested to encourage technological innovation and technology transfer worldwide. However, while numerous innovative technologies are invented, most of them remain latent and un-transferred. The comprehension of technical documents and the identification of appropriate technologies for given needs are challenging problems in technology transfer due to information asymmetry and information overload problems. There is a lack of common knowledge base that can reveal the technical details of technical documents and assist with the identification of suitable technologies. To bridge this gap, this research proposes to construct knowledge graph for facilitating technology transfer. A case study is conducted to show the construction of a patent knowledge graph and to illustrate its benefit to finding relevant patents, the most common and important form of technologies.
  • Item
    Combining Enterprise Knowledge Graph and News Sentiment Analysis for Stock Price Prediction
    ( 2019-01-08) Liu, Jue ; Lu, Zhuocheng ; DU, Wei
    Many state of the art methods analyze sentiments in news to predict stock price. When predicting stock price movement, the correlation between stocks is a factor that can’t be ignored because correlated stocks could cause co-movement. Traditional methods of measuring the correlation between stocks are mostly based on the similarity between corresponding stock price data, while ignoring the business relationships between companies, such as shareholding, cooperation and supply-customer relationships. To solve this problem, this paper proposes a new method to calculate the correlation by using the enterprise knowledge graph embedding that systematically considers various types of relationships between listed stocks. Further, we employ Gated Recurrent Unit (GRU) model to combine the correlated stocks’ news sentiment, the focal stock’s news sentiment and the focal stock’s quantitative features to predict the focal stock’s price movement. Results show that our method has an improvement of 8.1% compared with the traditional method.
  • Item
    Exploring Evaluation Factors and Framework for the Object of Automated Trading System
    ( 2019-01-08) Huang, Danya ; Cheng, Xusen ; Hou, tingting ; Liu, Kun ; Li, Chengyao
    Automated trading system (ATS) is a computer program that combines different trading rules to find optimal trading opportunities. The objects of ATS, which are financial assets, need evaluation because that is of great significance for stakeholders and market orders. From the perspectives of dealers, agents, external environment, and objects themselves, this study explored factors in evaluating and choosing the object of ATS. Based on design science research (DSR), we presented a preliminary evaluation framework and conducted semi-structured interviews with twelve trading participants engaged in different occupations. By analyzing the data collected, we validated eight factors from literatures and found four new factors and fifty-four sub-factors. Additionally, this paper developed a relationship model of factors. The results could be used in future work to explore and validate more evaluation factors by using data mining.