public static void GetSentimentandNouns()
{
positiveFeatures.Columns.Add("pNouns", typeof(string));
negativeFeatures.Columns.Add("nNouns", typeof(string));
props.setProperty("annotators", "tokenize, ssplit, pos, parse, sentiment");
//props.setProperty("annotators", "tokenize, ssplit, parse");
props.setProperty("ner.useSUTime", "0");
List <string> nouns = new List <string>();
// List<string> adjectives = new List<string>();
string line = "";
int sentiment;
// We should change current directory, so StanfordCoreNLP could find all the model files automatically
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(jarRoot);
var pipeline = new StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
DbWrappers wrap = new DbWrappers();
List <Product> products = wrap.GetAllReviewsAgainstProduct();
foreach (var prod in products)
{
foreach (var review in prod.ProductReviews)
{
Annotation annotation = pipeline.process(review.Review);
var Sentence = annotation.get(typeof(CoreAnnotations.SentencesAnnotation)) as ArrayList;
foreach (CoreMap sen in Sentence)
{
sentiment = GetSentiment(sen.ToString(), pipeline);
if (sentiment == 3 || sentiment == 4)
{
java.lang.Class treeClass = new edu.stanford.nlp.trees.TreeCoreAnnotations.TreeAnnotation().getClass();
edu.stanford.nlp.trees.Tree tree1 = (edu.stanford.nlp.trees.Tree)sen.get(treeClass);
string treeString = tree1.toString();
MatchCollection mc = Regex.Matches(treeString, @"(([N][N]\s\w*)|([N][N][P]\s\w*) |([N][P]\s\w*))");
foreach (Match m in mc)
{
string opinionPhrase = "";
string opinionString = m.ToString();
var phrase = opinionString.Split(' ');
foreach (var subPhrase in phrase)
{
if (!(subPhrase.Contains("NP")) && !(subPhrase.Contains("NNP")) && !(subPhrase.Contains("NNS")) && !(subPhrase.Contains("NN")))
{
String refinedSubPharse1 = subPhrase.Replace(")", string.Empty);
//Console.WriteLine(refinedSubPharse1);
string refinedSubPharse = refinedSubPharse1.ToLower();
if (refinedSubPharse.Contains("harddrive") || refinedSubPharse.Contains("memory") || refinedSubPharse.Contains("hdd") || refinedSubPharse.Contains("keyboard") || refinedSubPharse.Contains("processor") || refinedSubPharse.Contains("speed") || refinedSubPharse.Contains("operating system") || refinedSubPharse.Contains("display") || refinedSubPharse.Contains("graphics") || refinedSubPharse.Contains("price") || refinedSubPharse.Contains("money") || refinedSubPharse.Contains("weight") || refinedSubPharse.Contains("camera") || refinedSubPharse.Contains("dimension") || refinedSubPharse.Contains("screen") || refinedSubPharse.Contains("ram") || refinedSubPharse.Contains("hard disk") || refinedSubPharse.Contains("os") || refinedSubPharse.Contains("battery") || refinedSubPharse.Contains("performance") || refinedSubPharse.Contains("looks") || refinedSubPharse.Contains("hardware") || refinedSubPharse.Contains("speed") || refinedSubPharse.Contains("design"))
{
opinionPhrases.Add(refinedSubPharse);
DataRow newRow = positiveFeatures.NewRow();
newRow["pNouns"] = refinedSubPharse.ToString();
positiveFeatures.Rows.Add(newRow);
}
//opinionPhrase += refinedSubPharse + " ";
}
}
}
}//end sentiment if
else if (sentiment == 1 || sentiment == 0)
{
java.lang.Class treeClass = new edu.stanford.nlp.trees.TreeCoreAnnotations.TreeAnnotation().getClass();
edu.stanford.nlp.trees.Tree tree1 = (edu.stanford.nlp.trees.Tree)sen.get(treeClass);
string treeString = tree1.toString();
MatchCollection mc = Regex.Matches(treeString, @"(([N][N]\s\w*)|([N][N][P]\s\w*) |([N][P]\s\w*))");
foreach (Match m in mc)
{
string opinionPhrase = "";
string opinionString = m.ToString();
var phrase = opinionString.Split(' ');
foreach (var subPhrase in phrase)
{
if (!(subPhrase.Contains("NP")) && !(subPhrase.Contains("NNP")) && !(subPhrase.Contains("NNS")) && !(subPhrase.Contains("NN")))
{
String refinedSubPharse1 = subPhrase.Replace(")", string.Empty);
//Console.WriteLine(refinedSubPharse1);
string refinedSubPharse = refinedSubPharse1.ToLower();
if (refinedSubPharse.Contains("harddrive") || refinedSubPharse.Contains("memory") || refinedSubPharse.Contains("hdd") || refinedSubPharse.Contains("keyboard") || refinedSubPharse.Contains("processor") || refinedSubPharse.Contains("speed") || refinedSubPharse.Contains("operating system") || refinedSubPharse.Contains("display") || refinedSubPharse.Contains("graphics") || refinedSubPharse.Contains("price") || refinedSubPharse.Contains("money") || refinedSubPharse.Contains("weight") || refinedSubPharse.Contains("camera") || refinedSubPharse.Contains("dimension") || refinedSubPharse.Contains("screen") || refinedSubPharse.Contains("ram") || refinedSubPharse.Contains("hard disk") || refinedSubPharse.Contains("os") || refinedSubPharse.Contains("battery") || refinedSubPharse.Contains("performance") || refinedSubPharse.Contains("looks") || refinedSubPharse.Contains("hardware") || refinedSubPharse.Contains("speed") || refinedSubPharse.Contains("design") || refinedSubPharse.Contains("keys") || refinedSubPharse.Contains("ram") || refinedSubPharse.Contains("hard disk") || refinedSubPharse.Contains("os") || refinedSubPharse.Contains("battery") || refinedSubPharse.Contains("performance") || refinedSubPharse.Contains("looks") || refinedSubPharse.Contains("hardware") || refinedSubPharse.Contains("speed") || refinedSubPharse.Contains("design") || refinedSubPharse.Contains("quick"))
{
opinionPhrases.Add(refinedSubPharse);
DataRow newRow = negativeFeatures.NewRow();
newRow["nNouns"] = refinedSubPharse.ToString();
negativeFeatures.Rows.Add(newRow);
}
//opinionPhrase += refinedSubPharse + " ";
}
}
}
} //end sentiment if
}
} //end of review loop
Dictionary <string, int> featureDictionary = new Dictionary <string, int>();
var pFeatures = from row in positiveFeatures.AsEnumerable()
group row by row.Field <string>("pNouns") into grp
select new { key = grp.Key, cnt = grp.Count() };
foreach (var obj in pFeatures)
{
if (obj.key.Equals("mouse") || obj.key.Equals("pad") || obj.key.Equals("touchpad"))
{
featureDictionary["touchpad"] += obj.cnt;
}
else
{
featureDictionary[obj.key] = obj.cnt;
}
FeatureSentiment fs = new FeatureSentiment();
fs.Feature = obj.key;
fs.PId = prod.Id;
fs.Sentiment = 4;
fs.Count = obj.cnt;
wrap.AddFeatureSentiment(fs);
}
var nFeatures = from row in negativeFeatures.AsEnumerable()
group row by row.Field <string>("nNouns") into grp
select new { key = grp.Key, cnt = grp.Count() };
foreach (var obj in nFeatures)
{
FeatureSentiment fs = new FeatureSentiment();
fs.Feature = obj.key;
fs.PId = prod.Id;
fs.Sentiment = 1;
fs.Count = obj.cnt;
wrap.AddFeatureSentiment(fs);
}
}
}