会計データを学習して、仕訳の入力時に摘要の内容から勘定科目を予測してみる ref: http://qiita.com/mix_dvd/items/9c2ef5e6fcc390be067e

file0.txt

import pandas as pd

filename = "JDL出納帳-xxxx-xxxx-仕訳.csv"
df = pd.read_csv(filename, encoding="Shift-JIS", skiprows=3)

file1.txt

columns = ["摘要", "借方科目", "借方科目正式名称"]
df_counts = df[columns].dropna()

file10.txt

from sklearn.externals import joblib

joblib.dump(vect, 'data/vect.pkl')
joblib.dump(clf, 'data/clf.pkl')
df_rs.to_csv("data/code.csv")

file11.txt

import pandas as pd

filename = "data/code.csv"
df = pd.read_csv(filename, header=None)
df.index = df.pop(0)
df_rs = df.pop(1)

from sklearn.externals import joblib

clf = joblib.load('data/clf.pkl')
vect = joblib.load('data/vect.pkl')

file12.txt

from janome.tokenizer import Tokenizer

t = Tokenizer()
tests = [
    "高速道路利用料",
    "パソコン部品代",
    "切手代",
]

notes = []
for note in tests:
    tokens = t.tokenize(note)
    words = ""
    for token in tokens:
        words += " " + token.surface
    notes.append(words)

X = vect.transform(notes)

result = clf.predict(X)

for i in range(len(tests)):
    print(tests[i], "\t[",df_rs.loc[result[i]], "]")

file13.sh

高速道路利用料   [ 旅費交通 ]
パソコン部品代   [ 消耗品費 ]
切手代   [ 通信費 ]

file2.sh

$ pip install janome

file3.txt

from janome.tokenizer import Tokenizer

t = Tokenizer()

notes = []
for ix in df_counts.index:
    note = df_counts.ix[ix,"摘要"]
    tokens = t.tokenize(note.replace('　',' '))
    words = ""
    for token in tokens:
        words += " " + token.surface
    notes.append(words.replace(' \u3000', ''))

file4.txt

from sklearn.feature_extraction.text import TfidfVectorizer

vect = TfidfVectorizer()
vect.fit(notes)

X = vect.transform(notes)

file5.txt

y = df_counts.借方科目.as_matrix().astype("int").flatten()

file6.txt

from sklearn import cross_validation

test_size = 0.2
X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=test_size)

file7.txt

from sklearn.svm import LinearSVC

clf = LinearSVC(C=120.0, random_state=42)
clf.fit(X_train, y_train)

clf.score(X_test, y_test)

file8.txt

tests = [
    "高速道路利用料",
    "パソコン部品代",
    "切手代"
]

notes = []
for note in tests:
    tokens = t.tokenize(note)
    words = ""
    for token in tokens:
        words += " " + token.surface
    notes.append(words)

X = vect.transform(notes)

result = clf.predict(X)

df_rs = df_counts[["借方科目正式名称", "借方科目"]]
df_rs.index = df_counts["借方科目"].astype("int")
df_rs = df_rs[~df_rs.index.duplicated()]["借方科目正式名称"]

for i in range(len(tests)):
    print(tests[i], "\t[",df_rs.ix[result[i]], "]")

file9.sh

高速道路利用料   [ 旅費交通 ]
パソコン部品代   [ 消耗品費 ]
切手代   [ 通信費 ]