Can you provide the Json files , lcquad_answers.json

Hi,
I have a query if the above solution can be effectively leveraged for a custom-built dataset? If yes, can u share the pipeline to be followed to build a custom query based KG.
Thanks

Hi,
In the notebooks 2_entities_KBQA I got errors when trying to run this function

def evaluate_entity_ranking(_e_spans, indices, top_n):
    '''
    Estimate ranking accuracy:
    n_samples <int> size of the sample questions pool
    top_n <int> threshold for the number of top entities 
    '''
    n_correct_entities, n_correct_entities_1hop = 0, 0
    n_correct_answers_1hop = 0
    # match entities
    for i in indices:
        top_e_ids = []
        
        # entities index lookup
        for span in _e_spans[i]:
            for match in e_index.match_label(span, top=top_n):
                top_e_ids.append(match['_source']['id'])
        
        if set(correct_entities_ids[i]).issubset(set(top_e_ids)):
            n_correct_entities += 1
        
        # extract a subgraph for top entities
        kg = HDTDocument(hdt_path+hdt_file)
        # all predicates: 1 hop
        kg.configure_hops(1, [], namespace, True)
        entities, _, _ = kg.compute_hops(top_e_ids)
        if set(correct_entities_ids[i]).issubset(set(entities)):
            n_correct_entities_1hop += 1
        if set(correct_answers_ids[i]).issubset(set(entities)):
            n_correct_answers_1hop += 1
        kg.remove()

The HDTDocument class from hdt package doesn't have attributes configure_hops and compute_hops
Could you please provide me some information about those two methods?
Thank you.

Hello,

I'm trying to execute: "notebooks/1_question_parser.ipynb", but '1hop' and '2hop' are not exist in JSON files. I'd like to understand how to get or generate them.

Other question, how is possible to reproduce this article experiment:
http://www.vldb.org/pvldb/vol10/p565-cui.pdf

I could not understand project in overview, thanks a lot, for any help.

Thank you for your contribution， Can you provide the data set?
@svakulenk0

when I do this step, there is an issue like below figure.

Sincerely need help.

Hi! What is the procedure for step 4 'Index entities and predicates into ES'? util/index.py requires a entity-frequency file, is that something we're supposed to create from dbpedia2016-04en.hdt and then feed into it? Thank you.