def get_fitness_gradient_for_single_solution(summ):
sys_tf = get_TF(summ, vocab)
# ref_tf is globally defined
fit_grad = fitness_gradient(sys_tf, ref_tf)
return fit_grad
thematic_words = [i for i, j in freq.most_common(N)]
max_thematic = max([len(np.intersect1d(i, thematic_words)) for i in sents])
for Si in sents:
Si_thematic = np.intersect1d(Si, thematic_words)
F5.append(len(Si_thematic) / max_thematic)
# F6: Proper Noun
for Si in sents:
Si_propnouns = np.intersect1d(Si, propernoun)
F6.append(len(Si_propnouns) / len(Si))
# F7: Similarities Between Sentences
vocab = sorted(set(flat))
TF = get_TF(sents, vocab)
sim_SiSj = []
for i, Si in enumerate(TF):
temp = []
for j, Sj in enumerate(TF):
if i == j: continue
temp.append(cosine(Si, Sj))
sim_SiSj.append(sum(temp))
max_simSiSj = max(sim_SiSj)
for sim_Si in sim_SiSj:
F7.append(sim_Si / max_simSiSj)
# F8: Term Weight
TFIDF = get_TFIDF(sents, vocab)
feature_file.close()
doc_file = open('./1.clean/001.txt')
doc = doc_file.read()
doc_file.close()
doc_T, doc_label, doc_sents = sentences_from_document(doc)
doc_T = doc_T[2:]
document = {}
for i in range(0, len(doc_sents)):
document[i + 1] = doc_sents[i]
vocab = generate_vocab(doc_sents)
ref_tf = get_TF(reference, vocab)
#STEP 1 - Initialize Set of Solutions A as weight of text features,
#Set of Velocities V, stage counter k = 1 and set the different
#parameters of SSO
random.seed()
best_solution = []
population_size = 100
no_of_features = 8
iterations = 100
sentences_to_extract = 12
velocity_multiplier = 0.75
inertia = 0.56
velocity_limiter_ratio = 0.37