Package org.apache.stanbol.enhancer.servicesapi

Examples of org.apache.stanbol.enhancer.servicesapi.ContentItem

102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
        engine = null;
    }
   
    @Test
    public void testEngineDe() throws IOException, EngineException {
        ContentItem ci = ciFactory.createContentItem(new StringSource(de_text));
        Assert.assertNotNull(ci);
        AnalysedText at = atFactory.createAnalysedText(ci, ci.getBlob());
        Assert.assertNotNull(at);
        ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE, new PlainLiteralImpl("de")));
        Assert.assertEquals("de", EnhancementEngineHelper.getLanguage(ci));
       
        //Add some Tokens with POS annotations to test the usage of
        //existing POS annotations by the lemmatizer
        Token verbrachten = at.addToken(de_verbStart,de_verbStart+de_verb.length());
View Full Code Here
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
    @Before
    public void setupTest() throws IOException {
        //create a contentItem for the plain text used for testing
        InputStream is = FstLinkingEngineTest.class.getClassLoader().getResourceAsStream(TEST_TEXT_FILE);
        Assert.assertNotNull("Unable to load '"+TEST_TEXT_FILE+"' via classpath",is);
        ContentItem ci = cif.createContentItem(new StreamSource(is,"text/plain"));
        AnalysedText at = atf.createAnalysedText(ci, ci.getBlob());
        is.close();
        //parse the prepared NLP results and add it to the ContentItem
        is = FstLinkingEngineTest.class.getClassLoader().getResourceAsStream(TEST_TEXT_NLP_FILE);
        Assert.assertNotNull("Unable to load '"+TEST_TEXT_NLP_FILE+"' via classpath",is);
        AnalyzedTextParser.getDefaultInstance().parse(is, Charset.forName("UTF-8"), at);
        is.close();
        //set the language of the contentItem
        ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE,
            EN_LANGUAGE));
        //set the contentItem and also the content
        this.ci = ci;
        this.content = at.getText().toString();
    }
View Full Code Here
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
      return ciFactory.createContentItem(cs);
    }

    @Test
    public void testEnhancementInterfaces() throws Exception {
        ContentItem ci = createContentItem(SINGLE_SENTENCE_SOURCE);
        UriRef ciUri = new UriRef(ci.getUri().getUnicodeString());
        RdfEntityFactory factory = RdfEntityFactory.createInstance(ci.getMetadata());
        long start = System.currentTimeMillis();
        //create an Text Annotation representing an extracted Person
        TextAnnotation personAnnotation = factory.getProxy(
                createEnhancementURI(), TextAnnotation.class);
        personAnnotation.setCreator(TEST_ENHANCEMENT_ENGINE_URI);
        personAnnotation.setCreated(new Date());
        personAnnotation.setExtractedFrom(ciUri);
        personAnnotation.getDcType().add(new UriRef("http://www.example.org/cv/annotatation-types/text#Person"));
        personAnnotation.setConfidence(0.8);
        personAnnotation.setSelectedText("Patrick Marshall");
        personAnnotation.setStart(SINGLE_SENTENCE.indexOf(personAnnotation.getSelectedText()));
        personAnnotation.setEnd(personAnnotation.getStart()+personAnnotation.getSelectedText().length());
        personAnnotation.setSelectionContext(SINGLE_SENTENCE);

        //create an Text Annotation representing an extracted Location
        TextAnnotation locationAnnotation = factory.getProxy(
                createEnhancementURI(),    TextAnnotation.class);
        locationAnnotation.setCreator(TEST_ENHANCEMENT_ENGINE_URI);
        locationAnnotation.setCreated(new Date());
        locationAnnotation.setExtractedFrom(ciUri);
        locationAnnotation.getDcType().add(new UriRef("http://www.example.org/cv/annotatation-types/text#Location"));
        locationAnnotation.setConfidence(0.78);
        locationAnnotation.setSelectedText("New Zealand");
        locationAnnotation.setStart(SINGLE_SENTENCE.indexOf(locationAnnotation.getSelectedText()));
        locationAnnotation.setEnd(locationAnnotation.getStart()+locationAnnotation.getSelectedText().length());
        locationAnnotation.setSelectionContext(SINGLE_SENTENCE);

        //create an Text Annotation representing an extracted Organisation
        TextAnnotation orgAnnotation = factory.getProxy(
                createEnhancementURI(),    TextAnnotation.class);
        orgAnnotation.setCreator(TEST_ENHANCEMENT_ENGINE_URI);
        orgAnnotation.setCreated(new Date());
        orgAnnotation.setExtractedFrom(ciUri);
        orgAnnotation.getDcType().add(new UriRef("http://www.example.org/cv/annotatation-types/text#Organisation"));
        orgAnnotation.setConfidence(0.78);
        orgAnnotation.setSelectedText("University of Otago");
        orgAnnotation.setStart(SINGLE_SENTENCE.indexOf(orgAnnotation.getSelectedText()));
        orgAnnotation.setEnd(orgAnnotation.getStart()+orgAnnotation.getSelectedText().length());
        orgAnnotation.setSelectionContext(SINGLE_SENTENCE);

        // create an Entity Annotation for the person TextAnnotation
        EntityAnnotation patrickMarshall = factory.getProxy(
                createEnhancementURI(), EntityAnnotation.class);
        patrickMarshall.setCreator(TEST_ENHANCEMENT_ENGINE_URI);
        patrickMarshall.setCreated(new Date());
        patrickMarshall.setExtractedFrom(ciUri);
        patrickMarshall.getDcType().add(new UriRef("http://www.example.org/cv/annotatation-types/entity#Entity"));
        patrickMarshall.setConfidence(0.56);
        patrickMarshall.getRelations().add(personAnnotation);
        patrickMarshall.setEntityLabel("Patrick Marshall");
        patrickMarshall.setEntityReference(new UriRef("http://rdf.freebase.com/rdf/en/patrick_marshall"));
        patrickMarshall.getEntityTypes().addAll(Arrays.asList(
                        new UriRef("http://rdf.freebase.com/ns/people.person"),
                        new UriRef("http://rdf.freebase.com/ns/common.topic"),
                        new UriRef("http://rdf.freebase.com/ns/education.academic")));
        // and an other for New Zealand
        EntityAnnotation newZealand = factory.getProxy(
                createEnhancementURI(), EntityAnnotation.class);
        newZealand.setCreator(TEST_ENHANCEMENT_ENGINE_URI);
        newZealand.setCreated(new Date());
        newZealand.setExtractedFrom(ciUri);
        newZealand.getDcType().add(new UriRef("http://www.example.org/cv/annotatation-types/entity#Entity"));
        newZealand.setConfidence(0.98);
        newZealand.getRelations().add(locationAnnotation);
        newZealand.setEntityLabel("New Zealand");
        newZealand.setEntityReference(new UriRef("http://rdf.freebase.com/rdf/en/new_zealand"));
        newZealand.getEntityTypes().addAll(Arrays.asList(
                new UriRef("http://rdf.freebase.com/ns/location.location"),
                new UriRef("http://rdf.freebase.com/ns/common.topic"),
                new UriRef("http://rdf.freebase.com/ns/location.country")));

        // and an other option for New Zealand
        EntityAnnotation airNewZealand = factory.getProxy(
                createEnhancementURI(), EntityAnnotation.class);
        airNewZealand.setCreator(TEST_ENHANCEMENT_ENGINE_URI);
        airNewZealand.setCreated(new Date());
        airNewZealand.setExtractedFrom(ciUri);
        airNewZealand.getDcType().add(new UriRef("http://www.example.org/cv/annotatation-types/entity#Entity"));
        airNewZealand.setConfidence(0.36);
        airNewZealand.getRelations().add(locationAnnotation);
        airNewZealand.setEntityLabel("New Zealand");
        airNewZealand.setEntityReference(new UriRef("http://rdf.freebase.com/rdf/en/air_new_zealand"));
        airNewZealand.getEntityTypes().addAll(Arrays.asList(
                new UriRef("http://rdf.freebase.com/ns/business.sponsor"),
                new UriRef("http://rdf.freebase.com/ns/common.topic"),
                new UriRef("http://rdf.freebase.com/ns/travel.transport_operator"),
                new UriRef("http://rdf.freebase.com/ns/aviation.airline"),
                new UriRef("http://rdf.freebase.com/ns/aviation.aircraft_owner"),
                new UriRef("http://rdf.freebase.com/ns/business.employer"),
                new UriRef("http://rdf.freebase.com/ns/freebase.apps.hosts.com.appspot.acre.juggle.juggle"),
                new UriRef("http://rdf.freebase.com/ns/business.company")));
        System.out.println("creation time "+(System.currentTimeMillis()-start)+"ms");

        //now test the enhancement
        int numberOfTextAnnotations = checkAllTextAnnotations(ci.getMetadata());
        assertEquals(3, numberOfTextAnnotations);

        int numberOfEntityAnnotations = checkAllEntityAnnotations(ci.getMetadata());
        assertEquals(3, numberOfEntityAnnotations);
    }
View Full Code Here
107
108
109
110
111
112
113
114
115
116
117
    }

    @Test
    public void testHtml() throws EngineException, IOException {
        log.info(">>> testHtml <<<");
        ContentItem ci = createContentItem("test.html", "text/html; charset=UTF-8");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
137
138
139
140
141
142
143
144
145
146
147
    }
    @Test
    public void testPdf() throws EngineException, IOException {
        log.info(">>> testPdf <<<");
        //PDF created by Apple Pages
        ContentItem ci = createContentItem("test.pdf", "application/pdf");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
194
195
196
197
198
199
200
201
202
203
204
            "</body></html>");
    }
    @Test
    public void testMsWord() throws EngineException, IOException {
        log.info(">>> testMsWord <<<");
        ContentItem ci = createContentItem("test.doc", "application/msword");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
222
223
224
225
226
227
228
229
230
231
232
            "The Stanbol enhancer can detect famous cities",
            "</body></html>");    }
    @Test
    public void testRtf() throws EngineException, IOException {
        log.info(">>> testRtf <<<");
        ContentItem ci = createContentItem("test.rtf", "application/rtf");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
251
252
253
254
255
256
257
258
259
260
261
            "</body></html>");
    }
    @Test
    public void testOdt() throws EngineException, IOException {
        log.info(">>> testOdt <<<");
        ContentItem ci = createContentItem("test.odt", "application/vnd.oasis.opendocument.text");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
280
281
282
283
284
285
286
287
288
289
290
            "</body></html>");
    }
    @Test
    public void testEMail() throws EngineException, IOException, ParseException {
        log.info(">>> testEMail <<<");
        ContentItem ci = createContentItem("test.email.txt", "message/rfc822");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
338
339
340
341
342
343
344
345
346
347
348
       
    }
    @Test
    public void testMp3() throws EngineException, IOException, ParseException {
        log.info(">>> testMp3 <<<");
        ContentItem ci = createContentItem("testMP3id3v24.mp3", "audio/mpeg");
        assertFalse(engine.canEnhance(ci) == CANNOT_ENHANCE);
        engine.computeEnhancements(ci);
        Entry<UriRef,Blob> contentPart = ContentItemHelper.getBlob(ci,
            singleton("text/plain"));
        assertNotNull(contentPart);
View Full Code Here
TOP

Related Classes of org.apache.stanbol.enhancer.servicesapi.ContentItem

  • $.MultiEnhancer
  • org.apache.stanbol.cmsadapter.cmis.mapping.CMISContenthubFeeder
  • org.apache.stanbol.cmsadapter.jcr.mapping.JCRContenthubFeeder
  • org.apache.stanbol.commons.web.base.writers.JsonLdSerializerProviderTest
  • org.apache.stanbol.contenthub.search.featured.FeaturedSearchImpl
  • org.apache.stanbol.contenthub.store.clerezza.ClerezzaStore
  • org.apache.stanbol.contenthub.store.inmemory.InMemoryStore
  • org.apache.stanbol.contenthub.store.solr.SolrStoreImpl
  • org.apache.stanbol.contenthub.web.resources.StoreResource
  • org.apache.stanbol.enhancer.benchmark.impl.BenchmarkImpl
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.