产品评论挖掘研究综述

产品评论挖掘是以Web上用户发表的产品评论为挖掘对象,采用自然语言处理技术,从大量的文本数据中发现关于产品的功能和性能的评价信息的过程。产品评论挖掘是一个新兴的研究领域,是对自然语言描述的无结构数据进行数据挖掘的典型代表。产品评论中挖掘得到的信息不仅可以帮助生产厂商改进产品,还可以帮助用户合理的购买产品。对产品评论挖掘进行了全面深入地讨论,介绍了产品评论挖掘系统的通用框架,然后对产品特征提取、主观句定位、用户态度提取、态度极性判定、挖掘结果显示这5个子任务进行了详细地阐述,最后介绍了产品评论挖掘的最新方向。     

酒店在线评论数据的特征挖掘

论文以酒店在线评论数据为研究对象，对酒店在线评论数据的特征挖掘进行了研究。论文首先从酒店在线评论数据的获取出发，经过数据清洗、词性分析、特征抽取、指标确定、特征筛选、特征确定、特征校验几个环节，实现了酒店在线评论数据特征挖掘的目的。论文以词频为基础，融合了词性分析、聚类分析等方法，利用词频数（TF）、词频率（TF₁）、词频权重（TTW）、评论频率（DF）、逆文档频率（IDF）和TF1-IDF等指标对候选特征词进行降维，得出酒店在线评论数据的特征，并对特征词进行校验，完成了酒店在线评论数据的特征挖掘的过程。论文将为以评论为依据的客户分类、酒店分类、智能推荐奠定基础。     

基于故障诊断的数据挖掘算法研究

本论文以复杂工业过程控制系统为研究对象.首先根据工业过程数据库特点提出了相应的数据挖掘技术及其改进算法,而后通过对工业过程控制中的故障诊断方面的研究构建出了利用数据挖掘技术实现处理系统故障诊断的总体设计图,并通过实验验证了此系统能够实现在线监控与故障分析与诊断.     

基于观点挖掘的产品可用性建模与评价

提出基于观点挖掘的产品可用性建模与评价方法。以Web上的产品评论为数据,利用观点挖掘的方法从非结构化评论中抽取结构化数据,选取与可用性相关的产品特征,使用因子分析法提取影响产品可用性的公共因子,建立产品可用性模型。对产品可用性进行评价,结果表明,该方法可以有效地从用户角度发掘产品可用性中存在的缺陷,为产品设计提供依据。     

基于粗糙集理论的属性约简算法研究

数据挖掘是近年来数据库领域中出现的一个新兴研究热点,它是从大量数据中获取知识。进行数据挖掘的方法很多,粗糙集方法便是其中的主要方法之一。属性约简算法是基于粗糙集理论的数据挖掘模型中的关键步骤,同时也是粗糙集理论研究中的一个研究重点。通过对粗糙集理论的属性约简算法的深入研究,本文提出了一种改进的属性约简启发式算法。该算法建立在可辨识矩阵计算基础上。改进算法基于Hu的算法与Jelonek算法,在计算可辨识矩阵的基础上,保证最终能够找到决策信息系统的一个约简,同时较Jelonek算法相比,运算时间明显减少。     

基于细粒度观点挖掘和Kano模型的用户满意度分析研究

在线评论在客户关系管理、产品营销等方面发挥着重要作用，有效利用在线评论来分析用户满意度，对企业改善其服务和产品至关重要。传统的满意度分析方法的变量设计往往依赖专家建议，较少考虑正负属性的不对称影响。为解决这些问题，利用文本细粒度观点挖掘技术，对用户在线评论进行特征挖掘，构建产品服务质量分数，并采用PRCA技术对服务属性的正负影响进行量化，将服务属性投射为Kano属性分类，然后分析不同粒度下不同品牌的客户满意度特点，并给出不同品牌的属性优先顺序。最后，从咖啡评论数据中挖掘出5个关键属性。实验结果表明，不同属性对满意度影响具有不对称效应，且不同粒度下的顾客满意度影响因素具有不同的特点，并给出了相应的精细化企业管理策略。     

细粒度情感分析的酒店评论研究

酒店在线评论细粒度挖掘具有重要研究意义.以酒店在线评论具体特征属性和情感分类为研究目标,应用Apfiori算法和情感词典匹配算法,对重庆雾都宾馆在线评论数据深入挖掘,挖掘出用户最关注的酒店十大特征和满意度结果,进一步挖掘出商务出差等五种不同出游类型人最关注的酒店五大特征和满意度结果.这种方法不仅能对酒店领域评论进行分析,同样能够应用于其他领域.     

面向在线产品评论数据的有效性建模与测度研究

为提高对虚假评论的识别精度并对评论数据的有效性进行准确预测,提出一种面向在线产品数据的有效性建模与测量方法。通过结合基于核主成分的特征提取方法和最小二乘支持向量机对在线产品的虚假评论进行识别,另一方面,基于排序Logit构建回归模型对量化的评论数据进行有效性判别预测。实验结果表明,本文方法在虚假评论识别和数据有效性分析方面效果良好,可以为消费者提供了更为精确的消费参考、为商业机构提供了更具辨识意义的评论数据,具有良好的应用价值。     

一种对象完备度优先填补的决策树规则提取算法

不完备信息系统中决策规则的提取是数据挖掘领域的重要研究问题。对不完备信息系统中决策规则的主要获取方法进行分析,以决策属性具有缺失值的不完备决策表为研究对象,提出一种基于数据优先填补的决策树规则提取算法。针对ROUSTIDA算法在数据填补时运算量较大且容易导致决策规则冲突这一问题,算法采用决策属性优先填补的思想,引入对象完备度概念对其进行改进,使用改进的ROUSTIDA算法对不完备决策表进行一次性数据填补预处理,并在限制容差关系下采用属性重要性为启发函数构建决策树,从而获得决策规则。实例表明该方法是有效的,生成的决策规则简单,且具有较高的精确度。     

形式概念分析中的对象概念与属性概念

为了进一步提高数据表示和数据挖掘的效率,对两类特殊概念即对象概念和属性概念进行了研究。分析了对象概念和属性概念与不可约元的关系,提出了对象概念和属性概念的识别算法;提出了以属性概念为递归终止条件的计算内涵缩减递归算法;研究了属性排序以及属性序列在规则提取中的应用。     

The evaluation for perceived quality of products based on text mining and fuzzy comprehensive evaluation

With the growth of the Internet and electronic commerce, more and more customers browse online reviews to understand products and service reputation. Online reviews can provide decision support for customers to purchase a product that is to their satisfaction. Manufacturers can also mine and analyze valuable information in favor of design and production from online reviews. Customer satisfaction is mainly determined by perceived quality of products. Hence, this study establishes a new method to evaluate the perceived quality by combining text mining with a fuzzy comprehensive evaluation method. The new evaluation method offers ideas and methods for future work to combine text mining technology with traditional evaluation methods. Customers can also make better purchase decisions and manufacturers design and manufacture better products by using this evaluation method.     

Dynamical mining of ever-changing user requirements: A product design and improvement perspective

Previous studies carried out customer surveys by questionnaires to collect data for analyzing consumer requirements. In recent years, a large and growing body of literature has investigated the extraction of customer requirements and preferences from online reviews. However, since customer requirements change dynamically over time, traditional studies failed to obtain the change data of customer requirements and opinions based on sentiments expressed in reviews. In this paper, a new method for dynamically mining user requirements is proposed, which is used to analyze the changing behavior of product attributes and improve product design. Dynamic mining differs from the traditional need acquisition mainly in three aspects: (1) it involves dynamically mining user requirements over time (2) it adds changes in manufacturers’ opinions to the analysis (3) it allows for product improvement strategies based on the changing behavior of product attributes. First, text mining is adopted to collect customer and manufacturer review data for different time periods and extract product attributes. A Natural Language Processing tool is used to measure the importance weight and sentiment score of product attributes. Second, an approach for dynamically mining user requirements is introduced to classify product attributes and analyze the changes of attribute data in three categories over time. Finally, an improvement strategy for next-generation product design is developed based on the changing behavior of attributes. Moreover, a case study on vehicles based on online reviews was conducted to illustrate the proposed methodology. Our research suggests that the proposed approach can accurately mine customer requirements and lead to successful product improvement strategies for next-generation products.     

基于句法规则和HowNet的商品评论细粒度观点分析

该文提出一种基于句法规则和HowNet词典的商品评论细粒度观点分析方法,主要包括三个模块: 评价对象抽取、评价对象—评价词对抽取、评价对象总体观点得分计算。具体思路为: 首先,结合词性标注和频繁项集方法构建一个初始的评价对象词典,便于重用和修正商品的总体评价维度;其次,基于爬取的电商评论文本真实数据设计了评价对象—评价词对抽取规则;最后,借助HowNet词典分别计算不同评价维度的观点综合得分,进而对比同一商品不同品牌在各个维度下的总体观点评价,该方法在商品评论语料集上验证了有效性。     

Developing Playability Heuristics Based on Nouns and Adjectives from Online Game Reviews

The objective of this study is to develop playability heuristics based on a lexical analysis of nouns and adjectives frequently used in online game reviews. Built on a previous lexical analysis of adjectives in online game reviews, it is argued that nouns together with adjectives will likely provide more contextual information than adjectives alone, and therefore the patterns among these words can be used to develop playability heuristics. A revised lexical approach is adopted to analyze nouns and adjectives from 821,122 online reviews. Ninety seven factors emerge from this analysis. Based on the nouns and adjectives highly loaded on these factors, a new process is introduced and 90 playability heuristics are derived. This study significantly expands the current pool of playability heuristics that facilitate the computer game design process. The lexical method adopted in this study demonstrates its effectiveness in developing interface design guidelines based on a large number of online reviews on a system or product. It can be extended to other fields that are human-behavior centered.     

Dynamic modelling of customer preferences for product design using DENFIS and opinion mining

Previous studies mainly employed customer surveys to collect survey data for understanding customer preferences on products and developing customer preference models. In reality, customer preferences on products could change over time. Thus, the time series data of customer preferences under different time periods should be collected for the modelling of customer preferences. However, it is difficult to obtain the time series data based on customer surveys because of long survey time and substantial resources involved. In recent years, a large number of online customer reviews of products can be found on various websites, from which the time series data of customer preferences can be extracted easily. Some previous studies have attempted to analyse customer preferences on products based on online customer reviews. However, two issues were not addressed in previous studies which are the fuzziness of the sentiment expressed by customers existing in online reviews and the modelling of customer preferences based on the time series data obtained from online reviews. In this paper, a new methodology for dynamic modelling of customer preferences based on online customer reviews is proposed to address the two issues which mainly involves opinion mining and dynamic evolving neural-fuzzy inference system (DENFIS). Opinion mining is adopted to analyze online reviews and perform sentiment analysis on the reviews under different time periods. With the mined time series data and the product attribute settings of reviewed products, a DENFIS approach is introduced to perform the dynamic modelling of customer preferences. A case study is used to illustrate the proposed methodology. The results of validation tests indicate that the proposed DENFIS approach outperforms various adaptive neuro-fuzzy inference system (ANFIS) approaches in the dynamic modelling of customer preferences in terms of the mean relative error and variance of errors. In addition, the proposed DENFIS approach can provide both crisp and fuzzy outputs that cannot be realized by using existing ANFIS and conventional DENFIS approaches.     

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。     

Product concept evaluation and selection using data mining and domain ontology in a crowdsourcing environment

For product design and development, crowdsourcing shows huge potential for fostering creativity and has been regarded as one important approach to acquiring innovative concepts. Nevertheless, prior to the approach could be effectively implemented, the following challenges concerning crowdsourcing should be properly addressed: (1) burdensome concept review process to deal with a large amount of crowd-sourced design concepts; (2) insufficient consideration in integrating design knowledge and principles into existing data processing methods/algorithms for crowdsourcing; and (3) lack of a quantitative decision support process to identify better concepts. To tackle these problems, a product concept evaluation and selection approach, which comprises three modules, is proposed. These modules are respectively: (1) a data mining module to extract meaningful information from online crowd-sourced concepts; (2) a concept re-construction module to organize word tokens into a unified frame using domain ontology and extended design knowledge; and (3) a decision support module to select better concepts in a simplified manner. A pilot study on future PC (personal computer) design was conducted to demonstrate the proposed approach. The results show that the proposed approach is promising and may help to improve the concept review and evaluation efficiency; facilitate data processing using design knowledge; and enhance the reliability of concept selection decisions.     

How credible are online product reviews? The effects of self-generated and system-generated cues on source credibility evaluation

Online product reviews are important information sources in the consumer decision-making process. Despite the importance of online reviews in product evaluation, there is an emerging need to address the role of self-generated and system-generated information in enhancing the credibility of online reviews. Two experiments were conducted to examine the topic. Study 1 found significant interactions between perceived similarity and source reputation on the evaluation of trustworthiness and expertise. Study 2 extended the findings of study 1 by examining how argument quality influenced credibility perception under different levels of similarity and source reputation.