rule_classifier.py

import os
import sys
import re

# generate regular expression from a concept template
# the formate of template:
# concept \t father_concept \t keywords
def genConceptReg(file_concept="test.cpt", formate_col = 3):
	if not os.path.exists(file_concept):
		print("can not find concept template")
		os._exit(0)

	cased_regexes = []
	fin = open(file_concept,"r")
	for line in fin:
		line = line.rstrip("\n")
		if len(line.split("\t"))!= formate_col or line[0] == "#":
			continue
		info_list = line.split("\t")
		cased_regexes.append((info_list[2].rstrip("\r"), info_list[0], 0.9))
	fin.close()
	return cased_regexes


def classify(paper_text=None, cased_regexes = None, flag_cased = 1, threshold=0.5):
	ret = []
	if paper_text != None:
		for reg, tag, certainty in cased_regexes:
			if flag_cased == 1:  
				m = re.search(reg, paper_text)
			else:
				m = re.search(reg, paper_text,re.IGNORECASE)

			if m:
				ret.append((tag, certainty, 'Matched regex {}'.format(str(reg))))
	return ret