Hi Richard, I'm glad to hear from you and hope everything will be better with you. Thanks for you advice. Answer a simple question first, my undergraduate degree is also in Computer Science. I'm going to sleep now and I will expand my proposal according to your advice several hours later. I hope it is not very late for me to improve my proposal~ Best wishes to you and your family!
Yours Daniel On Thu, Apr 2, 2009 at 12:42 AM, Richard Frovarp <[email protected]>wrote: > Daniel Gong wrote: > >> Hi all, >> I'm a postgraduate student from Fudan University, Shanghai, China. >> This is my first time joining GSoC and I was not quite clear that I >> should exchange my ideas with possible mentors. I've submitted my proposal >> today. It's lucky that I can still modify it. >> Here is my proposal, any criticism and suggestions are welcome~ >> > Daniel, > > Sorry for taking so long to reply. Yes, working with mentors, and in fact > the community is part of the process, which is why it should go through the > dev list. > > ================================================ >> >> *Abstract: * >> >> The main idea dealing with the subject is to treat XML DOM structure as a >> DOM tree and translate the problem to computing diffs between tree >> structures. Some algorithms exist for tree diff computing, such as Tree Edit >> Distance. Some small modification should be made to adapt the algorithm to >> the context. >> >> *Detailed Description: * >> >> The implementation of the module can be divided into 4 parts: >> >> 1. Parse the XML text to get the DOM structure; >> 2. Translate the DOM structure to tree structure; >> 3. Employ some algorithm to computer the diffs; >> 4. Translate the tree diffs to XML diffs; >> 5. Display the diffs and maybe mail them. >> >> We would like to see some more detail in the detailed description. This > should be at least a couple of paragraphs long. It should show that you > understand the scope, goals, and awareness of issues you may have while > doing the project. > > 1. >> >> >> */Initial Algorithm Design/* >> >> According to my past research experience, Tree Edit Distance is a class of >> algorithms that using edit distance to measure tree similarity. The >> algorithms define 3 types of edit operations on labled tree: insert, delete >> and relabling. To measure the distance, the algorithms assign weights to >> operations, and define the minimum weight summary of all possible edit >> sequences between two trees as the edit distance.There is a corresponding >> best edit sequence with the minimum weight. The sequence can be translated >> to describe the diffs between XML texts. >> >> */Draft Timeline/* >> >> * Week 1 Complete a survey in the related area to decide the >> algorithm to employ; * Week 2-3 Implement the module of the XML >> parser and translater; >> * Week 4-6 Implement the algorithm chozen to compute tree diffs; >> * Week 7-8 Implement the module which translate the tree diffs to >> XML diffs and display them; >> * Week 9 Implement the module which can mail the diffs to certain >> mail address; >> * Week 10 Debug the whole module and make necessary modifications >> to successfully complete the subject. >> >> Good detail in the timeline. It is important to note in your proposal how > much time you have to dedicate towards the project and any other interfering > factors (tests, job, etc). > > * >> >> >> *Additional Information:* >> >> I've been learning and using Java since 3 years ago. Although my >> experience in dealing with XML text with Java is not that vast, my knowledge >> in programming, software architecture and algorithm can help me to learn >> fast and handle the problem. >> >> I'm 23 years old, living in Shanghai, China, attending Fudan University. >> >> ================================================ >> >> > You should further expand on your qualifications to do this job. What sort > of projects you have done in the past that are similar and show that you > have the skills to complete the task. For example, what is your > undergraduate degree in? > > Richard > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
