def create_feature_space_IG(n):
a={}
weight = {}
ig, gr, si = textgain.compute('IG')
sortedIG = sorted(ig,key=ig.__getitem__,reverse=True)
k = sortedIG[:n]
for j in range(0,len(k)):
if k[j] not in a:
a[k[j]] = len(a)
weight[k[j]] = ig[k[j]]
return [a,weight]
def create_feature_space_IG(n):
a = {}
weight = {}
ig, gr, si = textgain.compute('IG')
sortedIG = sorted(ig, key=ig.__getitem__, reverse=True)
k = sortedIG[:n]
for j in range(0, len(k)):
if k[j] not in a:
a[k[j]] = len(a)
weight[k[j]] = ig[k[j]]
return [a, weight]
def create_feature_space_IG(bi):
a = {}
ig, gr, si = textgain.compute('IG')
sortedIG = sorted(ig, key=ig.__getitem__, reverse=True)
ki = sortedIG[:8000]
k = bi[:16000]
k.extend(ki)
print len(k)
for j in range(0, len(k)):
if k[j] not in a:
a[k[j]] = len(a)
return a
def create_feature_space_IG(bi):
a={}
ig, gr, si = textgain.compute('IG')
sortedIG = sorted(ig,key=ig.__getitem__,reverse=True)
ki = sortedIG[:8000]
k = bi[:16000]
k.extend(ki)
print len(k)
for j in range(0,len(k)):
if k[j] not in a:
a[k[j]] = len(a)
return a