CS的SOP求建议！

*朋友的帐号借发

. From 1point 3acres bbs最近准备提交几个逼近Deadline的项目了....慌的一逼。担心有什么本人没看出来的缺陷。

由于我GPA太烂了，文书作业实在不能再承担太多失分了...所以想求求大家给个建议，批评指正一下！
(目前SOP的进度是给过母语人士润过稿，但就是纯润稿而已，没有内容上的建议)

以下內文：   
    I am applying for the Master of Computer Science (MCS) to pursue a career in practically implementing data analytics in real-world applications as a software engineer. To achieve my goal, I aim to improve my skills in three core areas: the efficiency of data management, solving problems with robustness, and hands-on real-world applications. My background in computer science, from undergraduate and master’s studies in Computer Science and Information Engineering (CSIE) at National Taiwan University (NTU), plus three years of industry experience, including internships in IBM and MediaTek and a full-time job in Cadence Design Systems, have lain the foundation for an MCS at the University of California, Irvine (UCI), where the courses and capstone projects will equip me with the skills I am seeking.

    My interest in data analytics started from the course “Information and Society” when I was a junior at NTU. I finished a project that aimed to help a volunteering club by evaluating how likely a certain activity was to attract a primary school student. I implemented logistic regression (written in C++ by myself) with raw data, which represented the first time I implemented an application for a real-life issue. The feedback from the club was positive, and they mentioned that my work helped them a lot in designing activities. Inspired by this experience of using computer science skills to contribute to society, I decided to continue my studies in computer science, and eventually applied to the master’s program at NTU.
    To further develop data analytics capabilities, I joined Prof. Liao’s lab and devoted myself to research analyzing financial data. Specifically, I constructed a portfolio evaluation method to help assess investment targets. Prior to my research, a common application of data analytics in finance was implementing good-at-handling-sequential-data models, like Hidden Markov Model (HMM) or long short-term memory (LSTM) network, with daily trading data as input features. However, trading data is usually not informative enough, since most fluctuations are manipulated or meaningless. Moreover, future prices can easily be influenced by external factors, such as accidents or big news, which cannot be predicted by previous trading data. Therefore, I preprocessed the raw data with a strategy-making algorithm called Online Moving Average Reversion algorithm (OLMAR), which adopts Passive Aggressive online learning (PA) to produce more moderate input data. Furthermore, for better performance, I collected textual data from the Internet with Python programs, and utilized sentiment analysis to consider the influence of information. As a result, my design achieved better performance, with 22% reduction in root mean square error (RMSE) and 35% improvement on sign prediction. My master’s research intensified my resolve to obtain more solid skills in implementing data analytics for real-world applications.

    Since finishing my studies in school, I have continued to enhance my skills and experience in my current role at Cadence Design Systems, which is one of the best-known electronic design automation (EDA) tool companies in the world. I work for the routing team, and our main task is to optimize the routing result (connection) of up to millions of electronic objects by deploying optimization models and graph algorithms. At Cadence, I have gotten experience contributing to large-scale software and improved my ability to collaborate with engineers from completely different fields. I have integrated a new feature into the routing flow by designing a weight function that will eventually reach a target ratio of 80%. I also help maintain an application, which needs a deep understanding of the workflow as well as frequent communications with different teams.
At Cadence, I have witnessed two problems: the efficiency issue in large design, and insufficient test cases because of patent rights. Since it is always critical to meet development deadlines, these problems not only lower the robustness of my work, but also prevent the possibility of introducing any long-training-time models into the routing stage of EDA. These observations have motivated me to apply for the MCS at UCI. I believe if I can improve my skills, then I might make it practical to implement data analytics in routing as well as other real-world applications.

    My participation in a volunteering club has honed my communication and teamwork skills, which will enable me to contribute to research groups during my master’s studies. During winter vacation, our club held a free winter camp for primary school students who lived in a remote area. The activities included free tutoring, arts and crafts, etc., and I was a team leader of a group of ten kids. Nonetheless, not until the beginning of the summer camp did I realize how difficult it is to manage primary school students. It was very frustrating. Yet, through the time I spent with them, I gradually understood how important it is to stand in others’ shoes. I listened to them, tried to understand their thoughts, and eventually learned to communicate with them. In my later study and work experience, my volunteer experience has always reminded me to foster open communication with my teammates, which has made me a better team player.

    I look forward to learning in the MCS at UCI because of the courses and capstone projects. I am particularly interested in taking “Principles of Data Management”, since storage management can be one root cause of inefficiency in a design. I am also very attracted to Prof. Li’s research on big data management, and would be honored to seek his advice in related fields. Furthermore, because of the incompleteness of test cases I’ve seen during work, I feel it is extremely important to acquire sophisticated skills in designing and developing a solution with robustness. The course “Advanced Programming and Problem Solving” can train me to do this. Last but not least, the “Capstone Design Project for Computer Science” offers an opportunity to verify my own skills and concept by designing and developing an app, which distinguishes UCI MCS from other programs. For these reasons, I believe that the MCS program at UCI is the natural next step in my career path, and hope you will give me the privilege of admission.

其实并不长，如果没有要求的话，SOP在500-1600字之间，或者1.5 - 2 页之间都可以。关于SOP怎么写，你可以看看这个帖子。https://www.1point3acres.com/bbs/thread-564958-1-1.html。顺便求个大米哈

我是读传媒的，不是读计算机的，所以不太清楚计算机的SOP是不是大部分都是楼主这个样式……
但从阅读感受来讲，有点“什么都讲”但并没有突出重点，只是从大三对XX感兴趣到后面加入实验室做了啥等等，梳理了一个有点流水账的东西。（可能由于专业差异，我的理解不够透彻，但不知为什么总觉得目前这个行文有点像把简历的内容复述了一遍，比如做了XXX，达到了什么结果，这个叙述就很像简历）
或许可以突出一两个重要的点，然后写写这个经历如何让你产生想要深造的想法，打算在深造时达成什么样的目标（展开具体一些讲，而不是开头几句话就泛泛讲过）。
另外，那段志愿者经历有点跳脱……
（以上内容很可能只是在胡说八道。楼主要不领英找个师兄师姐校友之类的帮忙看看？）

我也是刚开始写文书，一些个人意见（不一定对，欢迎交流～）。我觉得lz的文书逻辑挺清晰的，有自己清晰的target，并且每段经历的逻辑都可以串起来。如果内容缩减一些就更好了，我自己在写经历的时候也会容易多写，但是也根据 针对什么任务-遇到了什么问题/难点-如何解决的-有什么learning 的逻辑来写，把多余的话去掉，可以删减很多冗余内容。PS：同为工作党看到lz的文书收获很大，希望lz一切顺利～

太长了吧……超过一些学校的上限了吧

建议主要讲自己上过的课和科研经历以及未来的规划。

同意楼上说的，文书太长了，可以把志愿者这段删掉。
还有说一个我个人的意见，不代表这个意见一定正确。就是我觉得最后一句的privilege of admission不太好。。。

Z_LQ 发表于 2020-12-11 01:19
我是读传媒的，不是读计算机的，所以不太清楚计算机的SOP是不是大部分都是楼主这个样式……
但从阅读感受 ...

有请一个在美国工作的同学帮我review过一次，感觉我经历描述的部分确实有点wordy，压缩到了我动机陈述的篇幅。但我已经删减了150个字左右了:'(，可能精简陈述方面有点缺乏天分。

涂哈哈 发表于 2022-10-10 01:36
我也是刚开始写文书，一些个人意见（不一定对，欢迎交流～）。我觉得lz的文书逻辑挺清晰的，有自己清晰的ta ...

同工作党，大佬一起交流下吗~？

纯纯纯纯 发表于 2022-11-13 17:04
同工作党，大佬一起交流下吗~？

可以呀，怎么联系你呢～