Introduction
It was divided three parts, problem background, clarification and restatement and our work. In the chapter on the problem background, I noticed that it not only mentioned the background of the problem, but also mentioned the literature review of solving this problem at present, and found out the shortcomings of the current solutions, so as to solve it in the text.
There are some excellent sentences in C2003717:" Specifically speaking, in Amazon's design, customers are allowed to choose a number from 1 to 5 to express their satisfaction level wih the product, write any text-based messages as reviews and freely vote for other reviews they considered helpful. This are the main data sources for related companies to gain insights into the markets in which they participate, the timing of that participation, and the potential success of product design feature choices."
In the above, the paragraph mentioned the background of the article and some rules of users' comments. I think it would be better to add some current problem analysis if I can find the relevant literature.
As for the clarification and restatement, it's like a restatement of the problem, but C 2003717 briefly describes the data sets given by the topic, and briefly describes the problems raised by the topic.The chapter called "our work" is like a problem analysis. It not only contains the author's analysis and solutions to each problem, but also contains a flow chart of problem analysis. That leads to another question: What are the differences between the summary and problem analysis? I compared the summary of this paper with the problem analysis, and got the following conclusions.
Problem analysis is relatively simple, and usually only includes the solution to the problem but not the calculated results, and the evaluation of the model will not be written, mainly including the processing methods and modeling process of various indicators and data. I think it would be better to add the reasons for using this model.
Assumptions and notations
I noticed that there are no assumptions and notations in this paper, but I think it is important to add some model assumptions and explanations to symbols. In the chapter of assumptions, it should contains the reasons of assumptions as followings.(It's from C2007707, because I think they have done it very well)
Further, I also learned that the notations should contains symbol, explanation and it's dimension. (It was learned in the mathematical modeling writing training course this Saturday!)
How to describe the data pre-processing and mining?\
1.In order to ensure that the analysis is reliable and truthful, we usually need to ensure the continuity and authenticity of the data. Since the data provided by the data website is not complete and some missing value exist, we use the following methods to fill in the data......
2.xxx is useless for our subsequent analysis and we simply delete the property.
3.For xxx, we replace all x with x.
4. Then we clean xxx by replacing xxx with xxx as much as possible, and eliminating the xxx for further text-based emotion analysis and keyword extraction.
5.In the end, we eliminate 214 unqualified items and leave 11424,1606 and 18784 items for hair dryer, microwave and pacifier, respectively.
The text data pre-processing also should be redefined as following: