[PYTHON]PLAGON.py

[PYTHON]PLAGON.py

#  ____  _        _    ____  ___  _   _ 
# |  _ \| |      / \  / ___|/ _ \| \ | |
# | |_) | |     / _ \| |  _| | | |  \| |
# |  __/| |___ / ___ \ |_| | |_| | |\  |
# |_|   |_____/_/   \_\____|\___/|_| \_|
# --------------------------------------

from os import listdir as os_listdir, path as os_path
# pip install -U scikit-learn
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity


def vectorize(Text):
    """
    """
    return TfidfVectorizer().fit_transform(Text).toarray()

def similarity(doc1, doc2):
    """
    """
    return cosine_similarity([doc1, doc2])

def loop_comparaison(file_x, text_vector_x, new_vectors, results):
    """
    """
    for file_y, text_vector_y in new_vectors:
        sim_score = similarity(text_vector_x, text_vector_y)[0][1]
        student_pair = sorted((file_x, file_y))
        score = (student_pair[0], student_pair[1],sim_score)
        results.add(score)
    return results

def plagon_core(list_of_files, contents_of_files):
    """
    """
    results = set()

    vectors = vectorize(contents_of_files)
    s_vectors = list(zip(list_of_files, vectors))

    for file_x, text_vector_x in s_vectors:
        new_vectors = s_vectors.copy()
        new_index = new_vectors.index((file_x, text_vector_x))
        del new_vectors[new_index]

        results = loop_comparaison(file_x, text_vector_x, new_vectors, results)

    return results


if __name__ == "__main__":
    dir_ = "/home/d4rk3r/ACTUALC/vagrant/PYTHON/github/test_plagiat"
    list_of_files = [os_path.join(dir_, doc) for doc in os_listdir(dir_) if doc.endswith('.txt')]
    contents_of_files =[open(File).read() for File in list_of_files]

    for data in plagon_core(list_of_files, contents_of_files):
        print(data)

# results:
# ['tt2.txt', 'tt3.txt'] => 0.050632398572142946 (5%)
# ['tt.txt', 'tt3.txt'] => 0.06448929199938869) (6%)
# ['tt.txt', 'tt2.txt'] => 0.7492151128400741) (74%)