def classifyZipcode(location_text):
classifier = recognition_heuristic.naivebayes(recognition_heuristic.getwords)
classifier.setdb('test_100.db')
classified_zipcode = classifier.classify(location_text)
zipcode = str(classified_zipcode[0])
confidence = str(classfied_zipcode[1])
#query MYSQL database and append to python list
cur = db.cursor()
cur.execute("SELECT latitude,longitude FROM sqlbook.zipcounty WHERE zipcode = %s", zipcode)
rows = cur.fetchall()
latitude = rows[0][0]
longitude = rows[0][1]
likely_zipcode = {"location_text": location_text, "zipcode": zipcode, "latitude": latitude, "longitude": longitude}
weather = get__weather_data.getWeather(likely_zipcode['zipcode'])
likely_zipcode.update(weather)
return likely_zipcode
# Licence: <your licence>
#-------------------------------------------------------------------------------
#!/usr/bin/env python
import recognition_heuristic
import MySQLdb
import string
def main():
pass
#Step 1: classify objects in some text with a likely location code
#use on unstructured text
classifier = recognition_heuristic.naivebayes(recognition_heuristic.getwords)
#use this function if given a dictionary or key/value structured storage of text
##classifier = recognition_heuristic.naivebayes(wordmatrixfeatures)
#Step 2: Set the database to store the training. Set the region to train
classifier.setdb('likely_zipcode.db')
state = 'CA'
#Step 3: Train the text with a location code
#initial test training already carried out
##recognition_heuristic.frequency_train(classifier)
##recognition_heuristic.impact_train(classifier)
