1. Introduction Postgraduate dissertations are the main basis for measuring whether the academic level of graduate students meets the standard of degree awarded. The review of graduate dissertations is an evaluation of whether the dissertations have reached the corresponding level. Although the comment is short, it is very important, and it involves whether you can graduate, get the corresponding degree, or get the corresponding degree with excellent evaluation. Therefore, it is the bounden sacred duty of every paper reviewer to write comments objectively and accurately and make appropriate evaluations on graduate thesis work. 2. Evaluation According to the comments of the dissertation, it is necessary to evaluate the papers submitted for review. The assessment is mainly based on the Regulations on Academic Degrees in People's Republic of China (PRC) (adopted at the 13th meeting of the Standing Committee of the Fifth NPC on February 1980). Because graduate degrees are divided into master's degrees and doctor's degrees, the requirements for papers are different in principle. To apply for a master's degree thesis, the main examination contents are: (1) mastering the solid basic theory and systematic professional knowledge of this subject; (2) have the ability to engage in scientific research or independently undertake professional and technical work. For applying for doctoral thesis, the main assessment is: (1) mastering the solid and extensive basic theory and systematic and in-depth professional knowledge of this discipline; (2) Having the ability to engage in scientific research independently; (3) making creative achievements in science or specialized technology. 3. Under the guidance of the tutor, the graduate students at the evaluation level have completed the research and writing of their dissertations, and generally need the following evaluation: (1) the evaluation of the tutor; (2) Peer expert review; (3) the defense Committee defends the defense; (4) Review by the Academic Degrees Committee. The first three links require written comments. Due to different angles and different orientations, comments are written in different ways. Especially the conclusion of the review, the author will explain it in detail later. Generally speaking, the reviewer of master's degree thesis should be an expert with the title of associate higher vocational college or above, and generally should be a master's tutor; The reviewers of doctoral dissertations should be experts in higher vocational colleges and above, and generally should be doctoral supervisors. Comments on graduate dissertations require accurate wording and objective evaluation. Whether it is positive or negative, it must be well-founded and convincing. A good evaluation can't add one more word, nor can it lose one word. 4. Evaluation content Generally speaking, the evaluation of a dissertation includes the following aspects: (1) The topic selection of the dissertation mainly depends on whether it belongs to the forefront of this discipline and whether the research work has theoretical significance and practical value. In China, most of the projects listed in the National 973 Project Plan, the 863 Project, the National Natural Science Foundation and the National Key Scientific and Technological Projects are at the forefront of scientific development, because they have been demonstrated by peer experts in the process of project establishment, which is the focus or focus of peer attention. (2) Literature review mainly depends on whether we have mastered the latest trends and progress of related research work in this discipline. Literature review has two aspects: quantity and quality. In terms of quantity, it mainly depends on whether the main documents related to this subject field are basically covered around the theme of the paper, and in terms of quality, it mainly depends on whether the representative papers and influential main documents in this subject field are summarized. For example, the literature of influential academic journals and conferences in this discipline and the works of influential scholars should be reflected in the literature review. More importantly, the dissertation review needs to be summarized by the author's literature review to see if it summarizes the overall trend, law and problems to be solved in this discipline and research direction. (3) Main work and innovation, academic significance and value, which are the core contents of dissertation evaluation. For doctoral dissertations, the research work should be innovative and systematic. Comments should not only list, summarize and summarize the main work and innovations of the submitted review papers, but also point out their academic value. In the writing of dissertation comments, a prominent problem is that in the reviewed graduate dissertations, the general author has summarized the main work and innovations of the dissertations, and the reviewers of the dissertations should summarize and evaluate the work and innovations of the dissertations from a higher height and a broader perspective. (4) Thesis writing Thesis writing is an important aspect of dissertation evaluation. It mainly examines whether the writing of the paper is clear and coherent, whether the experimental data can be used, and whether the inference is scientific and correct. Through the writing of the paper, we can also see whether the author has a meticulous work style, and the orderliness and systematicness of the paper are also the embodiment of whether the research work is systematic. (5) Problems and Suggestions Any research work, whether high-level or low-level, will inevitably have problems, which is beyond reproach and is well known. But generally speaking, for quite some time before, domestic academic circles did not dare to point out that too sharp problems would affect the overall evaluation of the paper, so they only made a positive evaluation and did not point out the problems existing in the research work. In recent years, in the evaluation of dissertations in some key universities, special columns with problems have been set up, and many experts have truthfully pointed out that they have played a very good role. In fact, objectively speaking, the relationship between the achievements and the existing problems of the paper should be analyzed in detail, and it cannot be generalized. The fact that there are many or big problems does not necessarily mean that there are no creative results or the results are of little value. (6) Opinions and conclusions Through the review of doctoral dissertations, opinions and conclusions should include or imply the following contents: a. Does the dissertation reflect the author's solid and extensive basic theory and systematic and in-depth professional knowledge in this discipline; B. Whether the research work of dissertation can reflect the author's ability to engage in scientific research independently; C whether the research work of the paper has achieved creative results in science or special technology. The final comment conclusion is related to the identity of the commentator. In my opinion, if you are a doctoral supervisor, the final accurate conclusion is that the thesis meets the requirements of doctoral thesis and agrees to be reviewed by peer experts. If it is a peer review expert, it is agreed (or suggested) to organize a thesis defense committee to arrange defense if the thesis submitted meets the requirements of doctoral thesis. As for the thesis defense committee, it is proposed that: by secret ballot, all the votes (X votes) passed the student's doctoral thesis defense, and agreed (or suggested) to award the doctoral degree (engineering). Or: by secret ballot of the defense committee, a unanimous vote (X votes) passed the defense of xx students' doctoral thesis, agreed to graduate, and agreed (or suggested) to award (engineering) a doctorate. 5. Example Appreciation The following is a dissertation comment, which I think is an example worthy of appreciation: covering the whole process of product design with computer-aided design technology is the main content of current CAD research. Traditional CAD technology focuses on the detailed design and drawing output of auxiliary products, so it has great limitations. In this paper, graphic unit is used as the carrier of product design information. By means of motion analysis, function mapping, variant design and related design, computer-aided design technology is fully integrated into the process of product conceptual design, and a series of creative research results have been obtained: 1. Part structure is divided into three levels: part, functional structure and gene unit, and gene unit is organized according to functional structure, which is beneficial to realize functional-based conceptual design of parts. 2. The extraction method of product skeleton unit is proposed. Through the replacement methods of skeleton units such as insertion, deletion, replacement, decomposition, integration, cloning and derivation, the variant design of products is carried out under the condition of keeping the function unchanged. Different from the traditional parametric design based on size, the above-mentioned deformation design can lead to the change of product structure, thus providing an effective CAD means for innovative design. The representation of skeleton elements fully reflects the constraint relationship between the structure and other structures in the product. On the premise of ensuring the order and consistency of all structural units in the product, it reduces the extra large amount of data, which is beneficial to the repeated consideration and modification of the design scheme in conceptual design. 3. In the associative design, five kinds of associative constraint models are summarized, which provide a design basis for automatically generating export units in the detailed design stage. 4. Based on graphic element replacement and iteration technology, a modular product information modeling prototype system is constructed. On this basis, MCADDS system is developed and successfully applied to the design of punching and shearing machine tools and the traditional system design of XJD point machine. 5. The paper is rich in content, clear in organization and complete in structure, especially the application of CAD technology to assist the variant design of products and the repeated modification of design scheme during the design process. This article is an excellent doctoral thesis, and it is recommended to submit a defense. 6. Conclusion From a certain point of view, the comments on graduate dissertations not only evaluate the research work of graduate dissertations, but also reflect the comprehensive level of reviewers. It not only reflects the academic level of the reviewers, but also reflects their writing style. It belongs to a kind of professional practical writing in applied writing, which is worth our study.