基于用户评论的APP软件评价方法研究与实现

发布时间：2018-05-07 22:27

本文选题：APP软件 + 软件评价　；参考：《昆明理工大学》2017年硕士论文

【摘要】：随着社会快速发展,移动智能终端发展也慢慢超过个人电脑,与之伴随出现APP软件爆发增长。应用商店的出现减少用户寻找APP软件的难度,人们在不知不觉中已经在享受其带来的便利。然而,应用商店对于APP软件审核不全面,劣质、仿冒、恶意的APP软件乘虚而入,给众多用户带来不必要的麻烦。所以,研究如何判断APP软件的优劣是具有重要意义的。用户通过发表评论来评价APP软件的好坏,因此通过用户评论可以分析出APP软件的好坏。但应用商店中APP软件的用户评论存在无约束性,这也给评价APP软件造成了阻碍。因此,研究APP软件用户评论是否有关也显得极其重要。APP软件不同于一般软件,针对用户的APP软件评价方法并不完善且存在着如下问题:(1)APP软件具有一般软件不具有的特点,不能完全使用一般软件评价方法对APP软件进行评价。(2)APP软件用户评论具有简短性、强倾向性,同时也具有随意性、自由性,这对APP软件评价带来困难。为了解决上述问题,本文提出了筛选APP软件用户评论的方法并提取描述APP软件某方面特征或用户情感表达的词,同时根据提取的词提出APP软件评价方法,最后基于筛选后的用户评论进行APP软件评价。在此期间,主要完成了如下工作:(1)针对APP软件用户评论自由性、随意性问题,基于朴素贝叶斯分类算法和词频统计方法建立了筛选APP软件用户评论方法,并获取APP软件的特征词库;(2)提出了一种针对APP软件的评价模型。(3)建立APP软件评价模型指标的度量方法。(4)利用灰色系统理论的灰色关联计算评价指标权重。(5)通过实验并验证方法的有效性。
[Abstract]:With the rapid development of society, the development of mobile intelligent terminals is also slowly surpassing that of personal computers, which is accompanied by the explosion of APP software. The advent of app stores makes it easier for users to find APP software, and people are already enjoying the convenience it brings. However, the application store does not audit the APP software comprehensively, inferior, fake, malicious APP software to take advantage of the false entry, to many users unnecessary trouble. Therefore, it is of great significance to study how to judge the merits and demerits of APP software. Users evaluate the quality of APP software by issuing comments, so they can analyze the quality of APP software by user comments. However, user reviews of APP software in the app store are non-binding, which also hinders the evaluation of APP software. Therefore, it is very important to study whether the user comments of APP software are relevant. App software is different from general software. The evaluation method of APP software for users is not perfect and the following problems exist. It is difficult to evaluate APP software by using general software evaluation method, which is short, strong tendency, random and free. In order to solve the above problems, this paper proposes a method to screen APP software user comments and extract words describing some aspects of APP software or user emotion expression. At the same time, according to the extracted words, a APP software evaluation method is proposed. Finally, the APP software is evaluated based on the filtered user comments. During this period, the main work is as follows: (1) aiming at the problem of APP software users' comment freedom and arbitrariness, this paper establishes a APP software user comment screening method based on naive Bayesian classification algorithm and word frequency statistics method. And get the characteristic lexicon of APP software. (2) this paper presents an evaluation model for APP software. (3) A measure method of establishing evaluation model index of APP software. 4) using grey relation of grey system theory to calculate the weight of evaluation index. 5) through real application, the evaluation index weight of APP software is calculated by using the grey relation theory. Verify and verify the effectiveness of the method.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.56

【参考文献】