1 package org.apache.lucene.analysis.id;
4 * Licensed to the Apache Software Foundation (ASF) under one or more
5 * contributor license agreements. See the NOTICE file distributed with
6 * this work for additional information regarding copyright ownership.
7 * The ASF licenses this file to You under the Apache License, Version 2.0
8 * (the "License"); you may not use this file except in compliance with
9 * the License. You may obtain a copy of the License at
11 * http://www.apache.org/licenses/LICENSE-2.0
13 * Unless required by applicable law or agreed to in writing, software
14 * distributed under the License is distributed on an "AS IS" BASIS,
15 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16 * See the License for the specific language governing permissions and
17 * limitations under the License.
20 import java.io.IOException;
21 import java.io.Reader;
23 import org.apache.lucene.analysis.Analyzer;
24 import org.apache.lucene.analysis.BaseTokenStreamTestCase;
25 import org.apache.lucene.analysis.KeywordTokenizer;
26 import org.apache.lucene.analysis.ReusableAnalyzerBase;
27 import org.apache.lucene.analysis.Tokenizer;
30 * Tests {@link IndonesianStemmer}
32 public class TestIndonesianStemmer extends BaseTokenStreamTestCase {
33 /* full stemming, no stopwords */
34 Analyzer a = new ReusableAnalyzerBase() {
36 public TokenStreamComponents createComponents(String fieldName, Reader reader) {
37 Tokenizer tokenizer = new KeywordTokenizer(reader);
38 return new TokenStreamComponents(tokenizer, new IndonesianStemFilter(tokenizer));
42 /** Some examples from the paper */
43 public void testExamples() throws IOException {
44 checkOneTerm(a, "bukukah", "buku");
45 checkOneTermReuse(a, "adalah", "ada");
46 checkOneTermReuse(a, "bukupun", "buku");
47 checkOneTermReuse(a, "bukuku", "buku");
48 checkOneTermReuse(a, "bukumu", "buku");
49 checkOneTermReuse(a, "bukunya", "buku");
50 checkOneTermReuse(a, "mengukur", "ukur");
51 checkOneTermReuse(a, "menyapu", "sapu");
52 checkOneTermReuse(a, "menduga", "duga");
53 checkOneTermReuse(a, "menuduh", "uduh");
54 checkOneTermReuse(a, "membaca", "baca");
55 checkOneTermReuse(a, "merusak", "rusak");
56 checkOneTermReuse(a, "pengukur", "ukur");
57 checkOneTermReuse(a, "penyapu", "sapu");
58 checkOneTermReuse(a, "penduga", "duga");
59 checkOneTermReuse(a, "pembaca", "baca");
60 checkOneTermReuse(a, "diukur", "ukur");
61 checkOneTermReuse(a, "tersapu", "sapu");
62 checkOneTermReuse(a, "kekasih", "kasih");
63 checkOneTermReuse(a, "berlari", "lari");
64 checkOneTermReuse(a, "belajar", "ajar");
65 checkOneTermReuse(a, "bekerja", "kerja");
66 checkOneTermReuse(a, "perjelas", "jelas");
67 checkOneTermReuse(a, "pelajar", "ajar");
68 checkOneTermReuse(a, "pekerja", "kerja");
69 checkOneTermReuse(a, "tarikkan", "tarik");
70 checkOneTermReuse(a, "ambilkan", "ambil");
71 checkOneTermReuse(a, "mengambilkan", "ambil");
72 checkOneTermReuse(a, "makanan", "makan");
73 checkOneTermReuse(a, "janjian", "janji");
74 checkOneTermReuse(a, "perjanjian", "janji");
75 checkOneTermReuse(a, "tandai", "tanda");
76 checkOneTermReuse(a, "dapati", "dapat");
77 checkOneTermReuse(a, "mendapati", "dapat");
78 checkOneTermReuse(a, "pantai", "panta");
81 /** Some detailed analysis examples (that might not be the best) */
82 public void testIRExamples() throws IOException {
83 checkOneTerm(a, "penyalahgunaan", "salahguna");
84 checkOneTermReuse(a, "menyalahgunakan", "salahguna");
85 checkOneTermReuse(a, "disalahgunakan", "salahguna");
87 checkOneTermReuse(a, "pertanggungjawaban", "tanggungjawab");
88 checkOneTermReuse(a, "mempertanggungjawabkan", "tanggungjawab");
89 checkOneTermReuse(a, "dipertanggungjawabkan", "tanggungjawab");
91 checkOneTermReuse(a, "pelaksanaan", "laksana");
92 checkOneTermReuse(a, "pelaksana", "laksana");
93 checkOneTermReuse(a, "melaksanakan", "laksana");
94 checkOneTermReuse(a, "dilaksanakan", "laksana");
96 checkOneTermReuse(a, "melibatkan", "libat");
97 checkOneTermReuse(a, "terlibat", "libat");
99 checkOneTermReuse(a, "penculikan", "culik");
100 checkOneTermReuse(a, "menculik", "culik");
101 checkOneTermReuse(a, "diculik", "culik");
102 checkOneTermReuse(a, "penculik", "culik");
104 checkOneTermReuse(a, "perubahan", "ubah");
105 checkOneTermReuse(a, "peledakan", "ledak");
106 checkOneTermReuse(a, "penanganan", "tangan");
107 checkOneTermReuse(a, "kepolisian", "polisi");
108 checkOneTermReuse(a, "kenaikan", "naik");
109 checkOneTermReuse(a, "bersenjata", "senjata");
110 checkOneTermReuse(a, "penyelewengan", "seleweng");
111 checkOneTermReuse(a, "kecelakaan", "celaka");
114 /* inflectional-only stemming */
115 Analyzer b = new ReusableAnalyzerBase() {
117 public TokenStreamComponents createComponents(String fieldName, Reader reader) {
118 Tokenizer tokenizer = new KeywordTokenizer(reader);
119 return new TokenStreamComponents(tokenizer, new IndonesianStemFilter(tokenizer, false));
123 /** Test stemming only inflectional suffixes */
124 public void testInflectionalOnly() throws IOException {
125 checkOneTerm(b, "bukunya", "buku");
126 checkOneTermReuse(b, "bukukah", "buku");
127 checkOneTermReuse(b, "bukunyakah", "buku");
128 checkOneTermReuse(b, "dibukukannya", "dibukukan");
131 public void testShouldntStem() throws IOException {
132 checkOneTerm(a, "bersenjata", "senjata");
133 checkOneTermReuse(a, "bukukah", "buku");
134 checkOneTermReuse(a, "gigi", "gigi");