public List<KeyValuePair<string, double>> RankedResults()
{
List<Document> documents = new List<Document>();
HashSet<string> dataSet = new HashSet<string>();
foreach (var result in results)
{
Document d = new Document(result);
documents.Add(d);
foreach (var term in d.tokens()){
dataSet.Add(term);
}
}
//Build Document Vectors
Dictionary<string, Vector> documentVectors = new Dictionary<string, Vector>();
foreach (var document in documents)
{
documentVectors.Add(document.ToString(), new Vector(dataSet, document));
}
//Build Query Vector
Query query = new Query(queryString);
Vector queryVector = new Vector(dataSet, query);
Dictionary<string, double> relevance = new Dictionary<string, double>();
foreach (var documentVector in documentVectors)
{
relevance.Add(documentVector.Key, Vector.GetSimilarityScore(queryVector, documentVector.Value));
}
//Sort result by most relevant
List<KeyValuePair<string, double>> myList = relevance.ToList();
return myList;
}
public static void add(Document document)
{
numOfDocuments += 1;
List<string> terms = document.tokens();
for (int i = 0; i < terms.Count; i++)
{
//term is already in index
if (index.ContainsKey(terms[i]))
{
Dictionary<string, List<int>> temp = index[terms[i]];
//term already exists in document
if (temp.ContainsKey(document.ToString()))
{
tfIndex[terms[i]][document.ToString()] += 1;//update tfIndex
temp[document.ToString()].Add(i);
index[terms[i]] = temp;
}
//first occurence in a document of term that already exists
else
{
dfIndex[terms[i]] += 1;
tfIndex[terms[i]][document.ToString()] = 1;
List<int> positionsList = new List<int>();
positionsList.Add(i);
temp[document.ToString()] = positionsList;
}
}
//new term entry
else
{
Dictionary<string, List<int>> temp = new Dictionary<string, List<int>>();
List<int> positionsList = new List<int>();
positionsList.Add(i);
temp[document.ToString()] = positionsList;
index[terms[i]] = temp;
Dictionary<string, int> tf = new Dictionary<string, int>();
tf[document.ToString()] = 1;
tfIndex[terms[i]] = tf;
dfIndex[terms[i]] = 1;
}
}
serialize();
}
static void Main(string[] args)
{
Document a = new Document("C:\\Users\\LOLU\\Documents\\csc322\\doc1", "txt");
Document b = new Document("C:\\Users\\LOLU\\Documents\\csc322\\doc2", "txt");
Document c = new Document("C:\\Users\\LOLU\\Books~Tutorials\\OSS2014.pdf");
Document d = new Document("C:\\Users\\LOLU\\Books~Tutorials\\codility lessons\\1-TimeComplexity.pdf");
Document e = new Document("C:\\Users\\LOLU\\Books~Tutorials\\codility lessons\\2-CountingElements.pdf");
Document f = new Document("C:\\Users\\LOLU\\Books~Tutorials\\codility lessons\\3-PrefixSums.pdf");
Document g = new Document("C:\\Users\\LOLU\\Documents\\csc322\\test", "html");
InvertedIndex.add(a);
InvertedIndex.add(b);
InvertedIndex.add(c);
/*
Query query = new Query("open source information");
Console.WriteLine(query.QueryType());
Console.WriteLine(query.tokens().Count);
foreach(var item in query.RankedResults()){
Console.WriteLine(item);
}*/
}
public Vector(HashSet<string> dataSet, Document document)
{
this.dataSet = dataSet;
vectorRep = document.GetVector(dataSet);
}