基于贝叶斯网络的在线商品评价质量评估

随着电子商务领域的迅速发展,在线商品评价规模日益庞大,评价质量参差不齐,用户难以筛选有用评价信息做出购买决策,因此如何有效识别高质量评价信息成为重要议题。以在线商品评价的有用性投票为基础定义评价质量,使用贝叶斯网络表示在线商品评价的相似性及不确定性,通过对在线商品评价信息进行多维度特征统计,构建在线商品评价质量评估模型,使用概率推理机制对在线商品评价质量进行分类预测,并给出评价质量分类置信度。在真实数据集上验证模型有效性及高效性。     

Das Rezept für die perfekte Rezension?

Online product reviews, originally intended to reduce consumers’ pre-purchase search and evaluation costs, have become so numerous that they are now themselves a source for information overload. To help consumers find high-quality reviews faster, review rankings based on consumers’ evaluations of their helpfulness were introduced. But many reviews are never evaluated and never ranked. Moreover, current helpfulness-based systems provide little or no advice to reviewers on how to write more helpful reviews. Average review quality and consumer search costs could be much improved if these issues were solved. This requires identifying the determinants of review helpfulness, which we carry out based on an adaption of Wang and Strong’s well-known data quality framework. Our empirical analysis shows that review helpfulness is influenced not only by single-review features but also by contextual factors expressing review value relative to all available reviews. Reviews for experiential goods differ systematically from reviews for utilitarian goods. Our findings, based on 27,104 reviews from Amazon.com across six product categories, form the basis for estimating preliminary helpfulness scores for unrated reviews and for developing interactive, personalized review writing support tools.     

Studying the consistency of star ratings and the complaints in 1 & 2-star user reviews for top free cross-platform Android and iOS apps

How users rate a mobile app via star ratings and user reviews is of utmost importance for the success of an app. Recent studies and surveys show that users rely heavily on star ratings and user reviews that are provided by other users, for deciding which app to download. However, understanding star ratings and user reviews is a complicated matter, since they are influenced by many factors such as the actual quality of the app and how the user perceives such quality relative to their expectations, which are in turn influenced by their prior experiences and expectations relative to other apps on the platform (e.g., iOS versus Android). Nevertheless, star ratings and user reviews provide developers with valuable information for improving the overall impression of their app. In an effort to expand their revenue and reach more users, app developers commonly build cross-platform apps, i.e., apps that are available on multiple platforms. As star ratings and user reviews are of such importance in the mobile app industry, it is essential for developers of cross-platform apps to maintain a consistent level of star ratings and user reviews for their apps across the various platforms on which they are available. In this paper, we investigate whether cross-platform apps achieve a consistent level of star ratings and user reviews. We manually identify 19 cross-platform apps and conduct an empirical study on their star ratings and user reviews. By manually tagging 9,902 1 & 2-star reviews of the studied cross-platform apps, we discover that the distribution of the frequency of complaint types varies across platforms. Finally, we study the negative impact ratio of complaint types and find that for some apps, users have higher expectations on one platform. All our proposed techniques and our methodologies are generic and can be used for any app. Our findings show that at least 79% of the studied cross-platform apps do not have consistent star ratings, which suggests that different quality assurance efforts need to be considered by developers for the different platforms that they wish to support.     

CSLabel: An Approach for Labelling Mobile App Reviews

Mobile apps (applications) have become a popular form of software, and the app reviews by users have become an important feedback resource. Users may raise some issues in their reviews when they use apps, such as a functional bug, a network lag, or a request for a feature. Understanding these issues can help developers to focus on users’ concerns, and help users to evaluate similar apps for download or purchase. However, we do not know which types of issues are raised in a review. Moreover, the amount of user reviews is huge and the nature of the reviews’ text is unstructured and informal. In this paper, we analyze 3 902 user reviews from 11 mobile apps in a Chinese app store — 360 Mobile Assistant, and uncover 17 issue types. Then, we propose an approach CSLabel that can label user reviews based on the raised issue types. CSLabel uses a cost-sensitive learning method to mitigate the effects of the imbalanced data, and optimizes the setting of the support vector machine (SVM) classifier’s kernel function. Results show that CSLabel can correctly label reviews with the precision of 66.5%, the recall of 69.8%, and the F₁ measure of 69.8%. In comparison with the state-of-the-art approach, CSLabel improves the precision by 14%, the recall by 30%, the F₁ measure by 22%. Finally, we apply our approach to two real scenarios: 1) we provide an overview of 1 076 786 user reviews from 1 100 apps in the 360 Mobile Assistant and 2) we find that some issue types have a negative correlation with users’ evaluation of apps.     

Empirical findings on ontology metrics

Ontologies are becoming the preferred way of representing, dealing and reasoning with large volumes of information in several domains. In consequence, the creation, evaluation and maintenance of ontologies has become an engineering process that needs to be managed and measured using sound and reliable methods. As part of any ontology engineering or revision process, metrics can play a role helping in identifying possible problems or incorrect use of ontology elements, along with providing a kind of quality assessment that complements reviews requiring expert inspection. However, in spite of the fact that there are a number of ontology metric proposals described in the literature, there is a lack of empirical studies that provide a basis for their interpretation. This paper reports a systematic exploration of existing ontology metrics from a large set of ontologies extracted from the Swoogle search engine. The OntoRank value used inside the Swoogle to rank search results is used as a contrast for some existing metrics. The results show that the existing proposed metrics evaluated in general do not seem to be indicators of that ranking, but they can be helpful to identify particular kinds of ontologies.     

Management of data communications networks

Although network management installations vary widely in content and complexity, they also perform some or all of three basic tech control functions: monitoring, to verify channel status and activity; diagnosis, to locate and identify malfunctions; and reconfiguration, to reroute around failures and to allocate resources. The paper reviews the many techniques available to assist in finding solutions to network management problems. These techniques range from simple manual operations to highly sophisticated microprocessor-based controllers and software routines.     

Moodle平台数据挖掘方法设计与实现

教育数据挖掘是一个新兴的研究方向。如何把存储在教育软件系统中的数据转变为有意义的信息,并为教育决策、优化教学过程服务,已成为大多数教育工作者所关注的内容。文中总结了当前教育数据挖掘的研究现状,介绍了一种基于Excel的简单数据挖掘方法。该方法利用模糊C均值聚类算法,对Moodle平台积累的日志数据进行分析,找出有相似学习行为的学生,为学习社区的小组划分和研究学习模式服务。实验表明,该方法能够更准确地对学生进行分类,而且操作更为简单、方便。     

'Sorting the wheat from the chaff': investigating overload in educational discussion systems

Abstract  This paper reports on research to investigate design features of asynchronous discussion systems for higher education. The research aims to identify features that increase the benefits of discussion systems and reduce the problems. The paper focuses on a major theme that emerged from interviews with learners and teachers: information overload.
Interview findings, together with literature and system reviews, were used to identify four areas of possible system enhancement aimed at alleviating overload: branched message threading; user recommendations of messages; message filtering; and a personal 'clippings' area for useful messages. The paper discusses these enhancements and their evaluation in a prototype discussion system used on a UK Open University course.
Data from the student evaluation indicated that students preferred branched threading to linear threading; they thought filtering and clippings were helpful in principle, although they did not use them very much; and they felt uncomfortable with the idea of recommending each others' messages. Students felt less overloaded in the second part of the course, which was when the system enhancements were made available.     

一种基于支持向量机和主题模型的评论分析方法

在移动应用软件中,用户评论是一种重要的用户反馈途径.用户可能提到一些移动应用使用中的问题,比如系统兼容性问题、应用崩溃等.随着移动应用软件的广泛流行,用户提供大量无结构化的反馈评论.为了从用户抱怨评论中提取有效信息,提出一种基于支持向量机和主题模型的评论分析方法RASL（review analysis method based on SVM and LDA）以帮助开发人员更好、更快地了解用户反馈.首先对移动应用的中、差评提取特征,然后使用支持向量机对评论进行多标签分类.随后使用LDA主题模型（latent dirichlet allocation）对各问题类型下的评论进行主题提取与代表句提取.从两个移动应用中爬取5 141条用户原始评论,并对这些评论分别用RASL方法和ASUM方法进行处理,得到两个新的文本.与经典方法ASUM相比,RASL方法的困惑度更低、可理解性更佳,包含更完整的原始评论信息,冗余信息也更少.     

基于多特征Bi-LSTM-CRF的影评人名识别研究

近年来电影行业蓬勃发展,相关的信息抽取和分析技术日益受到行业内的重视,其中对电影主创人物的分析尤为重要。而电影评论作为观影群体的主要反馈信息,具有重要的分析价值。如何从影评中自动抽取主创人名成为重要的基础工作。然而评论中观众对人物的称谓方式多样复杂,而且新电影的影评中往往存在大量人名未登录词,传统方法难以有效识别。针对影评的这些特点,该文提出一种基于多特征Bi-LSTM-CRF的影评人名识别方法。该方法通过利用外部人名语料和未标注影评提取字符级的特征,并采用Bi-LSTM-CRF模型进行人名字符序列标注。实验结果表明,该方法能够有效识别影评中的复杂称谓和人名未登录词,从而有效地抽取影评中的人名实体。     

基于Hash技术的重复性评论检测

随着互联网技术的迅速发展,论坛已经成为人们获取信息、发表言论的重要场所,但大量的重复评论已成为论坛舆情信息内容获取与监管系统中新的难题,因此对重复评论进行有效检测和消重就至关重要.针对重复评论在一定时间内具有数量大、密度高、内容相似度高的特点,提出了一种基于SHA-1技术的重复评论检测方法.该方法以句和段为粒度块计算评论Hash值,然后统计Hash表中相同的指纹数目以此判断评论之间的相似度.最后依据给出的相似度阈值检测评论是否为重复评论.实验结果表明,该方法可以对重复评论进行有效检测和消重,且优于传统方法.     

Dynamic aspect-based rating system and visualization

User Modeling and User-Adapted Interaction - With an increasing number of product reviews available online, it has become impractical for potential customers to perceive all the available reviews...     

Reading to make a decision or to reduce cognitive dissonance? The effect of selecting and reading online reviews from a post-decision context

This research challenges the presumption that reading online reviews solely affect reader attitudes toward a prospective decision. Instead, readers may strategically select and read reviews after a decision. This research advances and tests hypotheses proposing that: (1) post-decision readers select decision-supportive reviews earlier and more frequently; and (2) the reviews they read affect the subsequent cognitive dissonance they experienced. Two studies employed an original post-decision experiment and demonstrated that readers expressed biased review selection. Interestingly, the reviews they read affected and increased their cognitive dissonance. The findings contributed to understanding the complex dynamic of online reviews in a post-decision context.     

Guidelines for conducting systematic mapping studies in software engineering: An update

ContextSystematic mapping studies are used to structure a research area, while systematic reviews are focused on gathering and synthesizing evidence. The most recent guidelines for systematic mapping are from 2008. Since that time, many suggestions have been made of how to improve systematic literature reviews (SLRs). There is a need to evaluate how researchers conduct the process of systematic mapping and identify how the guidelines should be updated based on the lessons learned from the existing systematic maps and SLR guidelines.ObjectiveTo identify how the systematic mapping process is conducted (including search, study selection, analysis and presentation of data, etc.); to identify improvement potentials in conducting the systematic mapping process and updating the guidelines accordingly.MethodWe conducted a systematic mapping study of systematic maps, considering some practices of systematic review guidelines as well (in particular in relation to defining the search and to conduct a quality assessment).ResultsIn a large number of studies multiple guidelines are used and combined, which leads to different ways in conducting mapping studies. The reason for combining guidelines was that they differed in the recommendations given.ConclusionThe most frequently followed guidelines are not sufficient alone. Hence, there was a need to provide an update of how to conduct systematic mapping studies. New guidelines have been proposed consolidating existing findings.     

基于DOM的评论发现及抽取模型研究

Internet发展至今,除了静态的网页文本,目前出现较多的是基于动态文本的BBS评论、电子商务评论等.如何自动挖掘这些评论信息,变得日益重要.提出了一种新颖的评论算法进行评论信息的发现与抽取.采用了DOM技术对页面进行分块,同时结合信息熵的迭代计算技术实现了评论块的自动发现与抽取.     

Do online reviews affect product sales? The role of reviewer characteristics and temporal effects

Online product reviews provided by consumers who previously purchased products have become a major information source for consumers and marketers regarding product quality. This study extends previous research by conducting a more compelling test of the effect of online reviews on sales. In particular, we consider both quantitative and qualitative aspects of online reviews, such as reviewer quality, reviewer exposure, product coverage, and temporal effects. Using transaction cost economics and uncertainty reduction theories, this study adopts a portfolio approach to assess the effectiveness of the online review market. We show that consumers understand the value difference between favorable news and unfavorable news and respond accordingly. Furthermore, when consumers read online reviews, they pay attention not only to review scores but to other contextual information such as a reviewer’s reputation and reviewer exposure. The market responds more favorably to reviews written by reviewers with better reputation and higher exposure. Finally, we demonstrate that the impact of online reviews on sales diminishes over time. This suggests that firms need not provide incentives for customers to write reviews beyond a certain time period after products have been released.

神经网络——通信及其保密通信领域的新技术

神经网络是近年来发展迅速的一门新型学科。它具有自适应学习、大规模并行分布处理、联想记忆以及强容错性等功能特点,是90年代最具希望和潜力的智能科学,各发达国家均不惜工本致力于该项研究,使神经网络的应用领域得到了长足发展,其中通信及其保密通信即是神经网络的重要应用领域,并已成为神经网络应用研究中的一个热点。本文在概述神经网络基本概念及其模型基础上,综述了神经网络在通信及其保密通信领域中的应用现状,并展望了神经网络在通信及其保密通信中的应用前景。     

Model-driven eGovernment interoperability: A review of the state of the art

This paper reviews the state of the art in the area of enhancing eGovernment interoperability by using common models and/or ontologies. This area has currently become a very active research field. We identify and present a significant number (> 40) of relevant efforts. These initiatives are grouped into categories based on the owner, scope and modelling perspective of each project. We then focus on the cases that build generic and universal eGovernment representations and models. We analyse, evaluate and rate them using an additional set of criteria. We end up with conclusions and possible directions for the exploitation and usage of these models.     

Training strategies for visual inspection

Airworthiness of the civil fleet is dependent upon the ability of inspectors to identify conditions for repair. As this is primarily a visual activity, the issue of training for visual inspection is critical. The inspection training literature has little detailed advice in this area, so in response to this need, a series of simulator experiments was undertaken for the development of improved training programs. This article briefly reviews each experiment and summarizes the results. Key conclusions drawn from each experiment are then used to demonstrate how they can be applied to enhance training for inspection tasks. © 1997 John Wiley & Sons, Inc.