Data Rules: How Mapping Technology Drives Better Customer Service

Is your library market-driven, product-driven or customer-driven? New internet-based data and tools are now available for public libraries that enable you to be customer-driven—for your own unique service area. This article shows how three library systems are using these tools—focused on market segmentation data and geographic information system (GIS) technology—to deliver more relevant customer services. Each solution described in the article was developed in close collaboration between CIVICTechnologies and the library, involving research to understand the library's objectives and desired outcomes, and using the data needed to meet each library's unique challenge.     

Marketing Law Libraries: Strategies and Techniques in the Digital Age

SUMMARY

Marketing is no longer a sporadic activity undertaken on an ad hoc basis, but rather has become an integral component of every library's day-to-day operations. This article provides an overview of basic marketing principles and then examines effective marketing strategies and promotional techniques in an academic environment. While viewed within the context of the law school setting, a majority of the marketing activities discussed are equally applicable in other types of law libraries.     

The role of culture,competitiveness and economic performance in explaining academic performance: a global market analysis for international student segmentation

A nation's culture, competitiveness and economic performance explain academic performance. Partial Least Squares (PLS) testing of 2252 students shows culture affects competitiveness and academic performance. Culture and economic performance each explain 32%; competitiveness 36%. The model predicts academic performance when culture, competitiveness and economic performance vary. A three-tier market categorisation enhances academic performance.     

基于IM的智能导学机器人的研究与设计

针对腾讯QQ、微软MSN等即时通讯工具在辅助教学中的一些缺陷,文章研究和设计了基于IM的能全天候实时在线进行辅助答疑的智能导学机器人,并详细阐述了导学机器人的系统架构、总体功能设计及其自动分词、自然语言生成、推理机制、机器学习等关键技术和算法。具有智能的IM机器人以其独特的亲和力、互动性和真实感将在远程辅助教学方面更大地提高学生的兴趣和满意度。     

Using symbolic convergence theory to discern and segment motives for enrolling in professional master's degree programs

In a preliminary effort to stem a steep decline in graduate enrollments, a regional comprehensive university began a search for its perceived market advantages. Using the tenets of Bormann's (1985) Symbolic Convergence Theory, the investigation identified fantasy types involved in student decisions to enroll and, after surveying new students (N = 739) over a four year period, isolated two emergent rhetorical visions ～ “Better Program, Quality University,” and “Small Campus, Big Opportunity”‐that were shared, to differing degrees, by students in professionally diverse academic programs. Results are discussed in terms of what they suggest about promoting professional master's degree programs generally, and communication programs specifically, and what they reveal about the saliency of rhetorical visions.     

图像分割中的阈值法研究

图像分割是图像处理中的一个重要问题,也是一个经典难题。阈值法计算简单,具有较高的运算效率,是图像分割中广泛采用的方法,主要分为全局阈值法和局部阈值法。系统地讨论了阈值法在图像分割中的应用,给出了一个比较全面的综述,对进行图像分割研究的科研人员具有一定的启发作用。     

基于哈希算法的中文分词算法的改进

中文分词是中文信息处理一个重要的部分,一些应用不仅要准确率,速度也很重要,通过对已有算法的分析,特别是对快速分词算法的分析,本文提出了一种新的词典结构,并根据新的词典给出了新的分词算法,该算法不仅对词首字实现了哈希查找,对词余下的字也实现哈希查找。理论分析和实验结果表明,算法在速度和效率比现有的几种分词算法上有所提高。     

MODIS亚像元积雪覆盖率提取方法

积雪覆盖区域,在全球能量平衡中发挥着重要作用.目前较常用的积雪覆盖率算法主要是利用雪盖率与雪盖指数(NDSI)间的关系建立统一的回归模型,但这种方法在完全雪盖区及稀疏雪盖区的反演精度较差.尝试通过分段建立模型反演雪盖率,并利用ETM+数据对模型估算结果进行验证.结果表明,分段模型可以有效提取亚像元尺度的信息,并且对NDSI高值区的雪盖率反演有一定的改善.     

一种基于多重哈希词典和K-最短路径算法的中文粗分词方案研究

本文在已有研究基础上,针对中文粗分词,设计了多重哈希词典结构,以提高分词的词典匹配效率,同时基于删除算法改进了中科院ICTCLAS分词系统的K-最短路径搜索思想.最后,论文对所研究技术方案进行了系统实现.系统实验结果表明,对于大规模文本,论文所提出的粗分词方案体现出了很好的性能.     

一种快速中文分词词典机制

通过研究目前中文分词领域各类分词机制,注意到中文快速分词机制的关键在于对单双字词的识别,在这一思想下,提出了一种快速中文分词机制:双字词-长词哈希机制,通过提高单双字词的查询效率来实现对中文分词机制的改进.实验证明,该机制提高了中文文本分词的效率.