//Inverse document frequency
public static void InverseDocumentFrequency(List <Reuter> reuters, List <Reuter> result)
{
double howManyDocumentsContainkeyword = 0;
result.Clear();
for (int i = 0; i < reuters.Count; i++)
{
if (reuters.ElementAt(i).Places.Count != 1)
{
continue;
}
result.Add(new Reuter {
Places = reuters.ElementAt(i).Places, TextTemp = reuters.ElementAt(i).TextTemp
});
result.Last().TextTemp = result.Last().TextTemp.Replace(" ", " ");
result.Last().Text = result.Last().TextTemp.Split(' ', '\n', '\t').ToList();
FeatureExtractions.HowManyWordsExtractor(result.Last());
}
foreach (Reuter r in result)
{
r.VectorFeatures = r.VectorFeatures.OrderBy(x => x.Value)
.Take(10)
.ToDictionary(pair => pair.Key, pair => pair.Value);
}
for (int i = 0; i < result.Count; ++i)
{
for (int j = 0; j < result[i].VectorFeatures.Count; ++j)
{
foreach (Reuter r in result)
{
if (r.Text.Contains(result[i].VectorFeatures.Keys.ElementAt(j)))
{
howManyDocumentsContainkeyword++;
}
}
double tempDiff = (double)result.Count / howManyDocumentsContainkeyword;
result[i].VectorFeatures[result[i].VectorFeatures.Keys.ElementAt(j)] = Math.Log10(tempDiff);
howManyDocumentsContainkeyword = 0;
}
}
}
public static void InverseDocument(List <Reuter> reuters, List <Reuter> result)
{
FeatureExtractions.InverseDocumentFrequency(reuters, result);
}
public static void HowManyWords(List <Reuter> result)
{
FeatureExtractions.HowManyWordsExtractor(result.Last());
}