use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.
the class FuzzyTokenMatcherTest method TestFuzzyTokenMatcher1.
@Test
public void TestFuzzyTokenMatcher1() throws Exception {
String query = "Twelve Angry Men";
// The ratio of tokens that need to be matched
double threshold = 0.5;
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.DESCRIPTION);
Schema schema = Utils.addAttributeToSchema(TestConstants.SCHEMA_PEOPLE, RESULTS_ATTR);
List<Span> spanList1 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 5, 10, "angry", "Angry", 1));
IField[] fields1 = { new StringField("bruce"), new StringField("john Lee"), new IntegerField(46), new DoubleField(5.50), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-14-1970")), new TextField("Tall Angry"), new ListField<Span>(spanList1) };
List<Span> spanList2 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "Angry", 1));
IField[] fields2 = { new StringField("brad lie angelina"), new StringField("pitt"), new IntegerField(44), new DoubleField(6.10), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-12-1972")), new TextField("White Angry"), new ListField<Span>(spanList2) };
List<Span> spanList3 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 40, 45, "angry", "Angry", 8));
IField[] fields3 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<Span>(spanList3) };
List<Span> spanList4 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "angry", 1));
IField[] fields4 = { new StringField("Mary brown"), new StringField("Lake Forest"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Short angry"), new ListField<Span>(spanList4) };
Tuple tuple1 = new Tuple(schema, fields1);
Tuple tuple2 = new Tuple(schema, fields2);
Tuple tuple3 = new Tuple(schema, fields3);
Tuple tuple4 = new Tuple(schema, fields4);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
expectedResultList.add(tuple2);
expectedResultList.add(tuple3);
expectedResultList.add(tuple4);
List<Tuple> results = FuzzyTokenMatcherTestHelper.getQueryResults(PEOPLE_TABLE, query, threshold, attributeNames);
boolean contains = TestUtils.equals(expectedResultList, results);
Assert.assertTrue(contains);
}
use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.
the class KeywordConjunctionTest method testMatchingWithLimit.
@Test
public void testMatchingWithLimit() throws TextDBException, ParseException, java.text.ParseException {
String query = "angry";
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.FIRST_NAME);
attributeNames.add(TestConstants.LAST_NAME);
attributeNames.add(TestConstants.DESCRIPTION);
Attribute[] schemaAttributes = new Attribute[TestConstants.ATTRIBUTES_PEOPLE.length + 1];
for (int count = 0; count < schemaAttributes.length - 1; count++) {
schemaAttributes[count] = TestConstants.ATTRIBUTES_PEOPLE[count];
}
schemaAttributes[schemaAttributes.length - 1] = new Attribute(RESULTS, AttributeType.LIST);
List<Tuple> resultList = KeywordTestHelper.getQueryResults(PEOPLE_TABLE, query, attributeNames, conjunction, 3, 0);
List<Tuple> expectedList = new ArrayList<>();
Span span1 = new Span("description", 5, 10, "angry", "Angry", 1);
Span span2 = new Span("description", 6, 11, "angry", "Angry", 1);
Span span3 = new Span("description", 40, 45, "angry", "Angry", 8);
Span span4 = new Span("description", 6, 11, "angry", "angry", 1);
List<Span> list1 = new ArrayList<>();
list1.add(span1);
IField[] fields1 = { new StringField("bruce"), new StringField("john Lee"), new IntegerField(46), new DoubleField(5.50), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-14-1970")), new TextField("Tall Angry"), new ListField<>(list1) };
List<Span> list2 = new ArrayList<>();
list2.add(span2);
IField[] fields2 = { new StringField("brad lie angelina"), new StringField("pitt"), new IntegerField(44), new DoubleField(6.10), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-12-1972")), new TextField("White Angry"), new ListField<>(list2) };
List<Span> list3 = new ArrayList<>();
list3.add(span3);
IField[] fields3 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<>(list3) };
List<Span> list4 = new ArrayList<>();
list4.add(span4);
IField[] fields4 = { new StringField("Mary brown"), new StringField("Lake Forest"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Short angry"), new ListField<>(list4) };
Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields1);
Tuple tuple2 = new Tuple(new Schema(schemaAttributes), fields2);
Tuple tuple3 = new Tuple(new Schema(schemaAttributes), fields3);
Tuple tuple4 = new Tuple(new Schema(schemaAttributes), fields4);
expectedList.add(tuple1);
expectedList.add(tuple2);
expectedList.add(tuple3);
expectedList.add(tuple4);
Assert.assertEquals(expectedList.size(), 4);
Assert.assertEquals(resultList.size(), 3);
Assert.assertTrue(TestUtils.containsAll(expectedList, resultList));
}
use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.
the class KeywordPhraseTest method testCombinedSpanInMultipleFieldsQuery.
/**
* Verifies: getNextTuple should return Combined Span info for the phrase
*
* @throws Exception
*/
@Test
public void testCombinedSpanInMultipleFieldsQuery() throws Exception {
// Prepare Query
String query = "lin clooney";
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.FIRST_NAME);
attributeNames.add(TestConstants.LAST_NAME);
attributeNames.add(TestConstants.DESCRIPTION);
// Prepare expected result list
List<Span> list = new ArrayList<>();
Span span1 = new Span("lastName", 0, 11, "lin clooney", "lin clooney");
Span span2 = new Span("description", 0, 11, "lin clooney", "Lin Clooney");
Span span3 = new Span("description", 25, 36, "lin clooney", "lin clooney");
list.add(span1);
list.add(span2);
list.add(span3);
Attribute[] schemaAttributes = new Attribute[TestConstants.ATTRIBUTES_PEOPLE.length + 1];
for (int count = 0; count < schemaAttributes.length - 1; count++) {
schemaAttributes[count] = TestConstants.ATTRIBUTES_PEOPLE[count];
}
schemaAttributes[schemaAttributes.length - 1] = new Attribute(RESULTS, AttributeType.LIST);
IField[] fields1 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<>(list) };
Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields1);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
// Perform Query
List<Tuple> resultList = KeywordTestHelper.getQueryResults(PEOPLE_TABLE, query, attributeNames, phrase);
// Perform Check
boolean contains = TestUtils.equals(expectedResultList, resultList);
Assert.assertTrue(contains);
}
use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.
the class KeywordPhraseTest method testWordInMultipleFieldsQueryWithStopWords1.
/**
* Verifies: Query with Stop Words match corresponding phrases in the
* document
*
* @throws Exception
*/
@Test
public void testWordInMultipleFieldsQueryWithStopWords1() throws Exception {
// Prepare Query
String query = "lin and and angry";
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.FIRST_NAME);
attributeNames.add(TestConstants.LAST_NAME);
attributeNames.add(TestConstants.DESCRIPTION);
// Prepare expected result list
List<Span> list = new ArrayList<>();
Span span1 = new Span("description", 25, 45, "lin and and angry", "lin clooney is Angry");
list.add(span1);
Attribute[] schemaAttributes = new Attribute[TestConstants.ATTRIBUTES_PEOPLE.length + 1];
for (int count = 0; count < schemaAttributes.length - 1; count++) {
schemaAttributes[count] = TestConstants.ATTRIBUTES_PEOPLE[count];
}
schemaAttributes[schemaAttributes.length - 1] = new Attribute(RESULTS, AttributeType.LIST);
IField[] fields1 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<>(list) };
Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields1);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
// Perform Query
List<Tuple> resultList = KeywordTestHelper.getQueryResults(PEOPLE_TABLE, query, attributeNames, phrase);
// Perform Check
boolean contains = TestUtils.equals(expectedResultList, resultList);
Assert.assertTrue(contains);
}
use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.
the class KeywordPhraseTest method testWordInMultipleFieldsQueryWithStopWords5.
/**
* Verifies: Query with Stop Words match corresponding phrases in the
* document Used to cause exception sometimes if there is a space between
* words
*
* @throws Exception
* with Medline data
*/
@Test
public void testWordInMultipleFieldsQueryWithStopWords5() throws Exception {
// Prepare Query
String query = "gain weight";
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(keywordTestConstants.ABSTRACT);
// Prepare expected result list
List<Span> list = new ArrayList<>();
Span span1 = new Span(keywordTestConstants.ABSTRACT, 26, 37, "gain weight", "gain weight");
list.add(span1);
Attribute[] schemaAttributes = new Attribute[keywordTestConstants.ATTRIBUTES_MEDLINE.length + 1];
for (int count = 0; count < schemaAttributes.length - 1; count++) {
schemaAttributes[count] = keywordTestConstants.ATTRIBUTES_MEDLINE[count];
}
schemaAttributes[schemaAttributes.length - 1] = new Attribute(RESULTS, AttributeType.LIST);
IField[] fields = { new IntegerField(4566015), new TextField(""), new TextField("Significance of milk pH in newborn infants."), new TextField("V C Harrison, G Peat"), new StringField("4-5839 Dec 2, 1972"), new TextField("British medical journal"), new TextField(""), new TextField("Infant Nutritional Physiological Phenomena, Infant, Newborn, Milk"), new TextField("Bottle-fed infants do not gain weight as rapidly as breast-fed babies during the first week of life. This " + "weight lag can be corrected by the addition of a small amount of alkali (sodium bicarbonate or trometamol) to " + "the feeds. The alkali corrects the acidity of cow's milk which now assumes some of the properties of human breast " + "milk. It has a bacteriostatic effect on specific Escherichia coli in vitro, and in infants it produces a stool with" + " a preponderance of lactobacilli over E. coli organisms. When alkali is removed from the milk there is a decrease in" + " the weight of an infant and the stools contain excessive numbers of E. coli bacteria.A pH-corrected milk appears to" + " be more physiological than unaltered cow's milk and may provide some protection against gastroenteritis in early " + "life. Its bacteriostatic effect on specific E. coli may be of practical significance in feed preparations where " + "terminal sterilization and refrigeration are not available. The study was conducted during the week after birth, and " + "no conclusions are derived for older infants. The long-term effects of trometamol are unknown. No recommendation can " + "be given for the addition of sodium bicarbonate to milks containing a higher content of sodium."), new DoubleField(0.667832788), new ListField<>(list) };
Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
List<Tuple> results = KeywordTestHelper.getQueryResults(MEDLINE_TABLE, query, attributeNames, phrase);
// Perform Check
boolean contains = TestUtils.equals(expectedResultList, results);
Assert.assertTrue(contains);
}
Aggregations