-
Directory case_study_result stores all figures, networks and text results.
-
After opening .html files in browser, figures can be downlowded with .png format from the webpage.
-
Directory to_be_checked contains files that need to be corrected by Profs.
- case_process
- process_details
- case_process_improved
- Process raw json file to structural json file, generate case dict files
- Phrase Matcher, entity recognition class, entity ruler, tokenizer, spacy pipeline
- Token2vec
- common_build_csv_pd
- Generate all_cases.csv from csv_graph
- Adding plaintiff, defendant, lawyer information for each case
- case_to_graph_only_method
- Generate csv_graph from csv files in total_dict_result/modified.csv, which contains information from all case dict files
- case_to_pd
- Generate modified.csv from all case dict files
- case_process_improved
- basic_information
- case_to_graph_basic_information
- Descriptive information of cases(1st, 2nd), plaintiffs, defendants, procedures, penalty, object money, (and legal fee)
- MoransI
- Explore whether case amount is relevant to provinces
- caseNeo4j
- Generate data for neo4j use
- case_to_graph_basic_information
- plaintiffs
- activePlaintiffPatterns1
- Descriptive information about the first 200 active plaintiffs and their cases
- Still have problems to be solved
- 用neo4j分析活跃原告和其被告的关系,可以分析重要的活跃原告,重要的被告,以及原告-被告pair
- activePlaintiffPatterns2
- Analytical information about the first 200 active plaintiffs and their cases
- activePlaintiffPatterns3
- Analyse the reason why active plaintiffs exit
- activePlaintiffPatterns4
- community analysis about 200 active plaintiffs
- plaintiffLawyerCommunityVisual
- Visualise community with sigma.js
- G_3_2.gexf contains community of plaintiffs that cooperated with the same lawyer
- plaintiffCommunityAnalysis
- Prove the validity of community in graph G_3 in active_plaintiff_patterns4
- Analyse lawyers of communities
- plaintiffLawyer
- Descriptive and analytical information about cases with lawyers and were sued by the first 200 active plaintiffs
- activePlaintiffPatterns1
- process_details
- reasonExtract
- caseReasonExtract
- extract case reasons through reason keywords match, and then vote match times to find the primary reason
- LDA
- Find case reason categories and keywords in each category from "court held" contents
- resultDisplay
- Descriptive information for case reasons
- textcnn
- classify "court held" content into specific case reason with self-defined reason category
- caseReasonExtract
- case_crawler
- Lawsdata website crawler based on python Scrapy frame