Function entity_cellvalues_lsh

union/TUS/freebaseProcess.py:17–36

(folder_path = 'benchmark', yago_path = 'yago', file_name='yagoTypes.tsv', threshold=0.95)

#通过这个lsh索引分桶里面选一个实体作为这个桶里所有值匹配的实体，阈值设为0.9，然后建个字典，存起来，处理表的时候直接查

def entity_cellvalues_lsh(folder_path = 'benchmark', yago_path = 'yago', file_name='yagoTypes.tsv', threshold=0.95):
    #file_path = os.path.join(yago_path, file_name)
    lsh = MinHashLSH(threshold=0.95, num_perm=128)
    entity_set = set()
    # with open(file_path, 'r', encoding='utf-8') as file:
    #     for line in file:
    #         _, entity, relation, t = line.strip().split('t')
    #         entity = process_string(entity[1:-1])
    #         entity_set.add(entity)
    conn, cursor = get_yagoTypes_from_tsv()
    query = "SELECT distinct entity, taxonomy_class, entity_count FROM yagoTypejoinclasses"
    cursor.execute(query)

    # 获取查询结果
    results = cursor.fetchall()
    for row in results:
        minhash = MinHash(num_perm=128)
        minhash.update(row[0].encode('utf-8'))
        index = "+-*/".join(row)
        lsh.insert(index, minhash)


def match_yago_entity_from_yago_lsh(lsh, file_subset):