Progress

About

• Name: Zhipeng Zhao
• Mentor: Luca Virgili
• Proposal: Social Knowledge Graph: Employing SNA measures to Knowledge Graph

Project Progress

[17/05/2021]: Proposal acceptance and community bonding period started.(excited)

[28/05/2021]: Meet with my mentor. It was a bumpy ride (a frequent Internet problem) but we had a great time.

[03/06/2021]: Meet all the students and mentors in DBpedia online. Glad to see each other!

[08/06/2021]: struggling with endpoints installation these days:( and begin to read the two papers about social network.

[10/06/2021]:

• endpoints installation -> change for online query(too big to load)
• finish reading the 1st paper Knowledge Graph Exploration where are we and where are we going -> done with Xmind and have a better understanding of our track

[11/06/2021]: Meet with my mentor.

• Exchange ideas with my mentor
  • Use 3 or more groups of entities from which extract the ego-networks, e.g scientists, companies, actors. From each group, select 2 or 3 entities.Save the obtained ego-networks.
  • NLP could be integrated. Word Embeddings are used in some papers of the literature
  • evaluate the similarity between two different entities such as Einstein and Kobe Bryant? Can we compute this similarity through the distances between the related graphs?(Graph Edit Distance, NetSimile)

[16/06/2021]:

• finish getting the top 10 nodes by degree
  • data analysis
  • data cleaning
  • Is it necessary to distinguish between incoming and outgoing degrees?（For an ideal structure, two nodes should have an edge pointing to each other）

[18/06/2021]:

• update the program
• problem: Always time out when querying -> solution: may get the json file manually

[19/06/2021]:

• meet with mentor
• change the order of program and tell the order I will be away 2 weeks for final exam weeks.

[22/06/2021]:

• draw the ego graph of scientists and get the top-10 suggestions by degree
• the center entity could also be the object and this bugs need to be fixed.

[19/06/2021 - 01/07/2021]:

• away for the final exam weeks

[02/07/2021]:

• make the data cleaning more reasonable to get the good results.

[05/07/2021 - 09/07/2021]:

• back to gsoc from the exams and school work
• try several python package to do the visualization, try to find the most proper one
  • pyecharts
  • plotly
  • pyvis

[10/07/2021]:

• meet with my mentor

1. the network problems with my cloud sever when developing with jupyterlab -> Switch to colaboratory and maybe use google free server
2. discuss the visulization tools. compare their features, advantages and disadvantages and show the mentor with the example project programmed by these tools for this task -> choose the plotly
3. exchange the ideas with this task: we should not only focus on entity or the relationship. we should take both into consideration
4. state the following week plan: continue to do the visualization.

[17/07/2021]:

• meet with my mentor
• State the possible way to do the entity importance computation -> need more statics analysis to see the resluts

[21/07/2021]:

Complete the code writing, enter the test, check whether the method is valid.

[21/07/2021]:

• meet with my mentor and share the results with him. Describe the continuous iterative process of the method.

[30/07/2021]:

• finish the data cleaning and write a short essay about the idea
• share the docs above with mentor and wait for his feedbacks.