semeval_parser.py

import xml.etree.ElementTree as ET
import pandas as pd

tree = ET.parse('Restaurants_Train.xml')
root = tree.getroot()

labeled_reviews = []
for sentence in root.findall("sentence"):
    entry = {}
    if sentence.find("aspectTerms"):
        terms = [term.get("term") for term in sentence.find("aspectTerms").findall("aspectTerm")] 
    if sentence.find("aspectCategories"): 
        aspects = [aspect.get("category") for aspect in sentence.find("aspectCategories").findall("aspectCategory")]
    entry["text"], entry["terms"], entry["aspects"]= sentence[0].text, terms, aspects
    labeled_reviews.append(entry)

labeled_df = pd.DataFrame(labeled_reviews)