Most Down Articles

    Published in last 1 year | In last 2 years| In last 3 years| All| Most Downloaded in Recent Month | Most Downloaded in Recent Year|

    Please wait a minute...
    For Selected: Toggle Thumbnails
    Techniques for estimating click-through rates of Web advertisements: A survey
    JI Wen-di, WANG Xiao-ling, ZHOU Ao-ying
    Journal of East China Normal University(Natural Sc    2013, 2013 (3): 1-14.  
    Abstract5808)      PDF (1639KB)(7509)      
    Computational advertising is a kind of advertising mechanism which has the capability to find the most suitable ads for given users and web content, so as to advertises them accurately. Therefore, estimating click-through rate (CTR) precisely makes significant difference in the efficiency of advertising on the Internet. Ad click-through rate prediction is to estimate CTR with click log, which is influenced by the nature features of ad, the position, the page information, user properties, the reputation of advertisers and such other factors. This paper is aimed to illustrate useful CTR prediction models, including CTR models for ads of abundant history data, CTR models for rare ads or new ads and some optimization models. Finally, the implementation methods with real data set were demonstrated as examples.
    Reference | Related Articles | Metrics
    Click-through rate prediction of online advertisements based on probabilistic graphical model
    YUE Kun, WANG Chao-lu, ZHU Yun-lei, WU Hao, LIU Wei-yi
    Journal of East China Normal University(Natural Sc    2013, 2013 (3): 15-25.  
    Abstract5405)      PDF (1294KB)(6866)      
    CTR (Click-Through Rate) prediction can be used to improve users’ satisfaction with respect to the presented online advertisements (ads) and support effective advertising. CTR prediction is the basis for personalized recommendation of online ads. It is also necessary to re-commend ads and predict their CTRs for the users that have no historical click-through records. In this paper, we adopted BN (Bayesian network), an important probabilistic graphical model, as the framework for representing and inferring the similarity and the corresponding uncertainty of the behaviors in ad search of different users. First, we constructed the BN to reflect the similarity between users by means of statistic computations on the historical records of user’s ad search. Then, we measured the behavior similarity between the users with click-through records and those without records quantitatively based on the mechanism of BN’s probabilistic inferences. Consequently, we predicted the CTRs of ads with respect to the users without historical click-through records, in order to provide a metric for ad recommendation. We made experiments on the training data of Tencent CA from KDD Cup 2012-Track 2 and tested the effectiveness of our methods.
    Reference | Related Articles | Metrics
    Ant colony optimization algorithm for computing resource allocation based on cloud computing environment (Chinese)
    HUA Xia-yu;ZHENG Jun;HU Wen-xin
    Journal of East China Normal University(Natural Sc    2010, 2010 (1): 127-134.  
    Abstract4502)      PDF (1353KB)(5970)      
    A new allocation algorithm based on Ant Colony Optimization (ACO) was established to satisfy the property of cloud computing. When start, this algorithm first prognosticated the capability of the potential available resource nodes, then analyzed some factors such as network qualities or response times to acquire a set of optimal compute resources. This algorithm met the needs of cloud computing more than others for grid environment with shorter response time and better performance, which were proved by the simulation results in the Gridsim environment.
    Related Articles | Metrics
    Online advertising systems and related technology evolution
    SONG Le-yi, GONG Xue-qing, ZHANG Rong, LIU Peng
    Journal of East China Normal University(Natural Sc    2013, 2013 (3): 106-117.  
    Abstract4398)      PDF (1590KB)(5863)      
    This paper introduced the background and related techniques of online advertising market. While different types of online ads may imply different computational advertising techniques in the backend system, this paper provided a rational classification method of online advertising formats. The category of online advertising given in this paper covers existing and potential types in the domain. Further, the advertising platform have evolved in several major phases or generations, particularly ad server, ad network and ad exchange. We showed the system architectures of the advertising systems mentioned above, by discussing the main function modules and interfaces. Our work aims to give a comprehensive and detailed description of the online advertising systems from the view of computation. Besides, the surveys in our work can provide essential background knowledge for computational advertising related research.
    Reference | Related Articles | Metrics
    Survey of online advertising target
    GUO Xin-yu, LIU Peng, ZHOU Min-qi, ZHOU Ao-ying
    Journal of East China Normal University(Natural Sc    2013, 2013 (3): 93-105.  
    Abstract3135)      PDF (1139KB)(5372)      
    Online advertising has sprung up and shown its diversity for decades. With the rapid increase of online advertising markets,a growing numbers of advertisers want to deliver ads related to their products or services to specific users with less expenditure. They expect these users have great interest in their products or services and have good intersection with ads(users see the ads, click it, registration or buy the products). Thus the concept of “advertising target” is proposed. This chapter discusses several classification of advertising target and for each category we elaborate on its application scenarios. We then summary and contrast the important advertising target methods and models used in recent years.
    Reference | Related Articles | Metrics
    Research of large scale graph data management with in memory computing techniques
    YUAN Pei-Sen, SHU Xin, SHA Chao-Feng, XU Huan-Liang
    Journal of East China Normal University(Natural Sc    2014, 2014 (5): 55-71.   DOI: 10.3969/j.issn.10005641.2014.05.005
    Abstract1715)      PDF (2216KB)(5315)      
    Graph is an important data model, which can describe structural information including dependent relationship, such as transportation network, social network and webpage hyperlink. Management of big graph brings challenges for traditional techniques, however, distributed cluster provide platform and techniques for this problem. Nowadays, the ratio of performance and price of memory promote rapidly, while demand of applications of highperformance, inmemory computing for massive data management is becoming popular. The storage and evaluation of massive graph requires highperformance platform. In this context, it’s significant for studying graph data management with inmemory techniques. This paper surveyes key techniques of management of massive graph data, and researched graph data management of inmemory computing techniques,and finally summarizes the paper.
    Related Articles | Metrics
    Research status and development studies of motion sensing technology
    ZHANG Shi-chao, QIAN Dong-ming
    Journal of East China Normal University(Natural Sc    2014, 2014 (2): 40-49, 126.  
    Abstract1933)      PDF (1021KB)(5294)      
    This thesis introduced the basic concepts and development of the human-computer interaction technology, as well as its applications in education and related areas. On this basis, Kinect would be used as a case study, including its application in the data of the depth, and skeleton tracking.
    Reference | Related Articles | Metrics
    Development of parallel computing models in the big data era
    PAN Wei, LI Zhan-Huai
    Journal of East China Normal University(Natural Sc    2014, 2014 (5): 43-54.   DOI: 10.3969/j.issn.10005641.2014.05.004
    Abstract1545)      PDF (459KB)(4876)      
    In the era of big data, the changing of the constraints gives the parallel computing opportunities and challenges for developing. This paper reviewed the new progress and changes of the parallel computing; combining with the effects of the hardware environments, computing pattern, application requirements on the parallel computing, the relevant research on batchoriented parallel computing model, streamingoriented parallel computing model, graphoriented parallel computing model and inmemory parallel computing model are summarized; Finally, the future development trends are evaluated.
    Related Articles | Metrics
    Simulator for hybrid memory architecture
    LIU Dong, ZHANG Jin-Bao, LIAO Xiao-Fei, JIN Hai
    Journal of East China Normal University(Natural Sc    2014, 2014 (5): 133-140.   DOI: 10.3969/j.issn.10005641.2014.05.011
    Abstract2043)      PDF (1508KB)(4874)      
    This paper proposed a method for building a simulator for hybrid memory architecture based on gem5. When building, this method first added a hybrid memory controller between the memory bus and the memory model, then introduced the nonvolatile memory model of NVMain and hooked it up to the the newly added hybrid memory controller along with the native DRAM model of gem5. This method could achieve the goal of building a simulator for hybrid memory architecture, which was proved by the experiment results.
    Related Articles | Metrics
    Analysis of tidal characteristics of the tide gauges in the Changjiang Estuary
    YANG Zheng-dong, ZHU Jian-rong, WANG Biao, LIN Tang-yu
    Journal of East China Normal University(Natural Sc    2012, 2012 (3): 111-119.  
    Abstract6508)      PDF (1190KB)(4713)      
    Based on the measured water levels at Hengsha, Majiagang, Baozhen and Yonglongsha tide gauges in 2009, the tidal temporal and spatial variations, tidal constituents, tidal form and distortion at these gauges in the Changjiang Estuary were analyzed. There exist tidal daily inequalities that mainly occur during high tide levels, and are more significant during neap tides than spring tides in March and September while are more significant during spring tides than neap tides in June and December. The monthly maximum and minimum tidal ranges in each month at the tide gauges were given out statistically. The tidal range in the estuary varies monthly with two peaks: the maximum tidal range reaches the maximum values in March and September, and reaches the minimum values in June and December. The minimum tidal range reaches the maximum values in June and December, and reaches the minimum values in March and September. The tidal range in the South Branch reduced toward upstream due to the river discharge and friction. The tidal range at Yonglongsha tidal gauge is the largest among the 4 tidal gauges because the North Branch has small river discharge division ratio and funnel shape bathymetry. The tide is mainly composed by the 4 semi-diurnal tidal constituent (M 2S 2N 2K 2), 4 diurnal tidal constituent (K 1O 1P 1Q 1) and 3 shallow water tidal constituent (M 4MS 4M 6). The semi-diurnal tidal constituents M 2 and S 2 are the predominant ones, and the shallow water tidal constituents M 4 and MS 4 are apparent due to the shallow water in the estuary. The tidal form number in the South Branch is larger than 0.25, indicating where the tide is an irregular semi-diurnal tide type there, and is smaller than 0.25 in the North Branch, indicating where the tide is a regular semi-diurnal tide type. The tidal distortion coefficients in the 4 tidal gauges are all larger than 0.1, indicating that the tide at the tide gauges has significant distortion in the Changjiang Estuary, especially at Yonglongsha in the North Branch with tidal distortion coefficient 0.173.
    Reference | Related Articles | Metrics
    Survey of resource uniform management and scheduling in cluster
    LI Yong-Feng, ZHOU Min-Qi, HU Hua-Liang
    Journal of East China Normal University(Natural Sc    2014, 2014 (5): 17-30.   DOI: 10.3969/j.issn.10005641.2014.05.002
    Abstract1716)      PDF (1403KB)(4691)      
    With the rapid development of Internet and the coming of big data, resource management system, a thin resource sharing layer that enables sharing cluster across diverse cluster computing frameworks, by giving frameworks a common interface for accessing cluster resources. For powering both large Internet services and a growing number of dataintensive scientific applications, cluster computing framework will continue emerge, and no framework will be optimal for all applications. Therefore, multiplexing a cluster between frameworks makes significant difference. Deploying and running multiple frameworks in the same cluster, improves utilization and allowing applications to share access to large datasets that may be costly to replicate across clusters. This paper is aimed to illustrate current major techniques of resource management and scheduling in cluster, including resource representation model, resource allocation model and scheduling policy. Finally, current prominent solutions, which have been developed and used by many companies, will be demonstrated, and we then summary and contrast these solutions used in recent years.
    Related Articles | Metrics
    Research on the framework of specification for e-Textbook and e-Schoolbag
    WU Yong-he, ZHU Zhi-ting, HE Chao
    Journal of East China Normal University(Natural Sc    2012, 2012 (2): 70-80.  
    Abstract5220)      PDF (2795KB)(4650)      
    With the development of information technology, the e-Textbook and e-Schoolbag is received more concerns from the community, its promotion and popularization is becoming the trend. First, this paper analyzes the demand of e-Textbook and e-Schoolbag from perspective of research on framework of standards. Then it describes the framework of e-Textbook and e-Schoolbag, including conceptual model, system framework model, hierarchy diagram for e-Textbook and e-Schoolbag system and function model. At last, it gives architecture of standards for e-Textbook and e-Schoolbag and profiles of related standards for e-Textbook and e-schoolbag.
    Reference | Related Articles | Metrics
    Towards the next generation of mobile recommender systems
    SONG Le-yi, XIONG Hui, ZHANG Rong
    Journal of East China Normal University(Natural Sc    2013, 2013 (3): 37-45.  
    Abstract3824)      PDF (392KB)(4527)      
    Recommender systems aim to identify content of interest from overloaded information by exploiting the opinions of a community of users. Due to the complexity of spatial data and the unclear roles of context-aware information, developing personalized recommender systems in mobile and pervasive environments is more challenging than developing recommender systems from traditional domains. This paper introduced classic recommendation techniques and unique features in mobile recommender systems, as well as the challenges in mobile enviroment. Based on two cases, taxi driving route recommendation and personalized travel package recommendation, we formulated the mobile sequential recommendation (MSR) problem and constrained travel recommendation. Finally, we gave a brief solution of the mobile recommender problem respectively.
    Reference | Related Articles | Metrics
    Young type inequalities for matrices
    HU Xing-kai
    Journal of East China Normal University(Natural Sc    2012, 2012 (4): 12-17.  
    Abstract2955)      PDF (144KB)(4485)      
    First, some Young type inequalities for scalars were given. Then on the base of them, corresponding Young type inequalities for matrices were established.
    Reference | Related Articles | Metrics
    Numerical simulations of effects on urban PBL characters with landuse categeories modification
    ZHANG Chi, SHU Jiong
    Journal of East China Normal University(Natural Sc    2011, 2011 (4): 83-93.  
    Abstract4139)      PDF (1102KB)(4158)      
    ENVI, IDL, GIS and aerial photographs were used to modify landuse data in ARW-WRF within Shanghai area. Firstly, grids were gained with GIS into the same resolution with the input data in WRF, which is too old for present landuse situation in Shanghai. Secondly, irrigated cropland and pasture, grassland, shrubland, loamy sand and silt were partly changed into urban and built-up land grid by grid, so as to be more closer to real situation, and the distribution of buildings was gained, helping to reach more accurate roughness value in UCM according to the 24 landuse categories defined in WRF model. Lastly, both modified and original data were input into WRF in case simulations. The wind field, temperature, and other characters of PBL in spring of Shanghai were gained in the simulation results. It shows that because of landuse types modification, horizontal wind velocity decreases obviously, displaying the urban dragging effects. Meanwhile, the vertical wind velocity increases, being greatly affected by surface warming effects. The higher temperature centers appear in the downwind direction by the effects of surface wind fields. Surface temperature and PBL height are both closer to observation values. The results lead us to discuss the importance of landuse categories input in models, and some other possible reasons causing the difference were mentioned.
    Reference | Related Articles | Metrics
    Improved collaborative filtering algorithm based on usersimilarity
    WANG Wei, ZHENG Jun
    Journal of East China Normal University(Natural Sc    2016, 2016 (3): 60-66.   DOI: 10.3969/j.issn.1000-5641.2016.03.007
    Abstract1224)   HTML184)    PDF (1219KB)(4113)      
    Collaborative filtering is widely accepted and applied currently as one of the most popular personalized recommendation methods. It is an implementation method based on content that has considerable advantages in accuracy. The core issue of collaborative filtering is how to work out the calculation of similarity. In this paper, we introduce the traditional collaborative filtering algorithm and make similarity calculation more accurately by optimizing the traditional formula of similarity. Experimental results show that the optimized algorithm can improve the accuracy of the recommendation and reduce the MAE (Mean Absolute Error, MAE) efficiently.
    Reference | Related Articles | Metrics
    Antiproliferative and apoptotic activities of novel curcumin analogs in human liver cancer cell lines
    LI Yu-bo, WEN Ying, MA Ming-liang, WU Liang-chun, WEN Ke, ZHAO Zheng
    Journal of East China Normal University(Natural Sc    2012, 2012 (3): 161-170.  
    Abstract2561)      PDF (2235KB)(4084)      
    Antiproliverative and apoptotic activities of the novel curcumin analogs (CCM series) against human liver carcinoma Bal-7402 and SMMC-7721 cells were investigated by MTT assay. The cell cycle distribution and apoptosis of SMMC-7721 cells induced by CCM-5 and CCM-14 were analyzed using flow cytometry. The expressions of caspase-3 and its activated form p17 in SMMC-7721 cells were further determined by western blot. CCM-5 and CCM-14 exhibited, in a concentration-dependent manner, the stronger antiproliferative role than those of curcumin and the other CCM compounds. Their apoptotic effects on the SMMC-7721 cells were also found to be significantly elevated as compared with the control group (P<0.01). Cell cycle distribution appeared that, as the concentrations of the compounds increased in SMMC-7721 cells, the G0/G1 phase cells decreased while the S phase and the G2/M phase cells, and the SubG1 peak increased. Furthermore, both CCM-5 and CCM-14 could activate caspase-3 expression in the SMMC-7721 cells. Collectively, our data suggest that CCM-5 and CCM-14 can restrain proliferation and promote apoptosis in SMMC-7721 cell, and the molecular mechanism underlying these actions against the cancer cells of the compounds may involve in the activation of caspase-3.
    Reference | Related Articles | Metrics
    Accurate eye location in near-infrared images based on ellipse fitting
    JIN Jun-cai, TONG Wei-qing, LIANG Xiao-ni, CHEN Qiang, MEI Yue-ping, LIU Dan
    Journal of East China Normal University(Natural Sc    2012, 2012 (3): 103-110.  
    Abstract2669)      PDF (3011KB)(4004)      
    This paper presents a novel approach to precisely locate eye position in near-infrared facial images. In this approach, we first determine the face region and initial eye position using face detection classifier based on Haar features and AdaBoost algorithm. Then we detect the eye edge in the eye region using Sobel operator, fit it into an elliptical contour. Finally, the center point of eye is located by the center of the fitted ellipse. With 120×120 normalized face images, the experiments show that the proposed approach is accurate. The average error is less than 1.5 pixels and the processing time is about 7 ms.
    Reference | Related Articles | Metrics
    Context aware computing(Chinese)
    GU Jun-zhong
    Journal of East China Normal University(Natural Sc    2009, 2009 (5): 1-20.  
    Abstract2982)      PDF (3326KB)(3956)      
    An overview of the basic concepts concerning context aware computing and a survey of the up-to-date researches were presented in this paper. Context and its spectrum were well formed and defined. Based on the evolution of computing modes, context aware computing as well as context aware systems was analyzed. As application examples, realizations of location based service and context aware web searching were discussed.
    Related Articles | Metrics
    Application of stationary technical indicator in high-frequency trading based on MACD
    BAO Si, ZHENG Wei-an, ZHOU Yu
    Journal of East China Normal University(Natural Sc    2013, 2013 (5): 152-160.  
    Abstract4116)      PDF (4857KB)(3931)      
    In recent years, the rapid development of the
    high-frequency trading in the global financial market causes the
    extensive concern of the financial world. Because of the ``high
    frequency'' character, the high-frequency trading cannot be
    implemented by manual operation, but only with the help of computer
    programming trading system. Therefore, building a reasonable model
    of high-frequency trading strategy is necessary. MACD is a very
    important and commonly used technical analysis indicator in the
    stocks, futures, foreign currency exchange market, it is commonly
    used to judge the buying or selling time, and track the running
    trend of price of assets. In this paper, we define a new stationary
    technical indicator $\widehat{\rm MACD}_t$   based on MACD
    indicator, which is suitable for high-frequency trading strategy
    modeling. We also prove the stationarity of $\widehat{\rm MACD}_t$
    under the hypothesis of stationarity of the increments of logarithm
    price process. Finally, we construct a high-frequency trading
    strategy based on $\widehat{\rm MACD}_t$  and test its effectiveness
    and profitability by using real market high-frequency data. All
    those put forward a new kind of thought in high-frequency trading.
    Reference | Related Articles | Metrics