KSR

Team Members

• Jan Karwowski
• Kamil Kowalewski

Task 1 - Console App Java SE

mvn clean package
java -jar <jar-with-dependencies> param1 param2 param3 param4 param5

Call parameters

* <percentage of training set (integer 1-99)> 
* <k for kNN (integer >0)> 
* <number of keywords (integer >0)>
* <numerical metric [eucl|manh|cheb]>
* <text metric [trigram|tfm]>

Extra params to remove selected feature:

* "1",
* "2,0-50", "2,50-100",
* "3,0-50,tfidf", 
* "3,50-100,tfidf", 
* "3,0-50,tfcitfoc", 
* "3,50-100,tfcitfoc",
* "4,0-50,tfidf", 
* "4,50-100,tfidf", 
* "4,0-50,tfcitfoc", 
* "4,50-100,tfcitfoc",
* "5,0-50,tfidf", 
* "5,50-100,tfidf", 
* "5,0-50,tfcitfoc", 
* "5,50-100,tfcitfoc",
* "6,0-50,tfidf", 
* "6,50-100,tfidf", 
* "6,0-50,tfcitfoc", 
* "6,50-100,tfcitfoc",

Script

Uncomment selected function in main function in main.py and run by python main.py
The rest of information is stored in script

Task 2 - JavaFX Desktop App + Spring Data JPA + PostgreSQL

Dataset

Database usage

• Install PostgreSQL with JDBC Driver
• Create database ksr_fuzzy manually or by using python script in sql-script directory
• Build JAR - mvn clean package and the run java -jar $JAR_NAME$ -sp to read data from csv file and save them into ksr_fuzzy database or restore database from data-dump.sql file after that run java -jar $JAR_NAME$ -sl to create and seed tables for labelWrapper and LinguisticQuantifierWrapper

Protip

• Install and create database ksr_fuzzy otherwise application with fail to start or comment out whole application.properties file - h2 database will be used

Call parameters

• Single param - seed_pollution or -sp to seed database from csv file
• Single param - seed_linguistic or -sl to seed database with linguistic data

Scipts description

• filter_data.py - filtering csv data - removes rows with NaN values, removes selected columns
• main.py - script with experiments for report
• merge_generated_summary.py - merging results from main.py
• remove_poor_summary.py - removing merged results from merge_generated_summary.py when T1 is below passed value
• draw_chart.py - generating chart for Quantifier
• generate_function.py - generating membership function formulas
• generate_function_to_file.py - usage of generate_function.py with selected params
• data-dump- script with dumped data from database after filtering - restore->custom or tar and filename dump-data

select no2aqi, date_local
from pollution 
where date_part('month', date_local)>=6
and date_part('month', date_local)<=9
order by no2aqi desc

Chosen screenshots