public void testHowMany()
{
IDataModel dataModel = getDataModel(
new long[] { 1, 2, 3, 4, 5 },
new Double?[][] {
new double?[] { 0.1, 0.2 },
new double?[] { 0.2, 0.3, 0.3, 0.6 },
new double?[] { 0.4, 0.4, 0.5, 0.9 },
new double?[] { 0.1, 0.4, 0.5, 0.8, 0.9, 1.0 },
new double?[] { 0.2, 0.3, 0.6, 0.7, 0.1, 0.2 },
});
IUserSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
IUserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, dataModel);
IRecommender recommender = new GenericUserBasedRecommender(dataModel, neighborhood, similarity);
IList <IRecommendedItem> fewRecommended = recommender.Recommend(1, 2);
IList <IRecommendedItem> moreRecommended = recommender.Recommend(1, 4);
for (int i = 0; i < fewRecommended.Count; i++)
{
Assert.AreEqual(fewRecommended[i].GetItemID(), moreRecommended[i].GetItemID());
}
recommender.Refresh(null);
for (int i = 0; i < fewRecommended.Count; i++)
{
Assert.AreEqual(fewRecommended[i].GetItemID(), moreRecommended[i].GetItemID());
}
}
static void GenericUserBasedRecommenderTest()
{
Console.WriteLine("GenericUserBasedRecommenderTest");
System.Diagnostics.Stopwatch watch = new System.Diagnostics.Stopwatch();
watch.Start();
var model = new FileDataModel(filePath);
var similarity = new PearsonCorrelationSimilarity(model);
var neighborhood = new NearestNUserNeighborhood(4, similarity, model);
var recommender = new GenericUserBasedRecommender(model, neighborhood, similarity);
var iter = model.getUserIDs();
while (iter.MoveNext())
{
var userId = iter.Current;
var recommendedItems = recommender.recommend(userId, 5);
Console.Write("uid:" + userId);
foreach (var ritem in recommendedItems)
{
Console.Write("(" + ritem.getItemID() + "," + ritem.getValue() + ")");
}
Console.WriteLine();
}
watch.Stop();
Console.WriteLine(watch.ElapsedMilliseconds);
}
private static IUserBasedRecommender buildRecommender()
{
IDataModel dataModel = getDataModel();
IUserSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
IUserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, dataModel);
return(new GenericUserBasedRecommender(dataModel, neighborhood, similarity));
}
public IRecommender BuildRecommender(IDataModel model)
{
IUserSimilarity similarity = new PearsonCorrelationSimilarity(model);
IUserNeighborhood neighborhood =
new NearestNUserNeighborhood(10, similarity, model);
return
(new GenericUserBasedRecommender(model, neighborhood, similarity));
}
public void testFile()
{
IUserSimilarity userSimilarity = new PearsonCorrelationSimilarity(model);
IUserNeighborhood neighborhood = new NearestNUserNeighborhood(3, userSimilarity, model);
IRecommender recommender = new GenericUserBasedRecommender(model, neighborhood, userSimilarity);
Assert.AreEqual(1, recommender.Recommend(123, 3).Count);
Assert.AreEqual(0, recommender.Recommend(234, 3).Count);
Assert.AreEqual(1, recommender.Recommend(345, 3).Count);
// Make sure this doesn't throw an exception
model.Refresh(null);
}
public void testIsolatedUser()
{
IDataModel dataModel = getDataModel(
new long[] { 1, 2, 3, 4 },
new Double?[][] {
new double?[] { 0.1, 0.2 },
new double?[] { 0.2, 0.3, 0.3, 0.6 },
new double?[] { 0.4, 0.4, 0.5, 0.9 },
new double?[] { null, null, null, null, 1.0 },
});
IUserSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
IUserNeighborhood neighborhood = new NearestNUserNeighborhood(3, similarity, dataModel);
IUserBasedRecommender recommender = new GenericUserBasedRecommender(dataModel, neighborhood, similarity);
long[] mostSimilar = recommender.MostSimilarUserIDs(4, 3);
Assert.NotNull(mostSimilar);
Assert.AreEqual(0, mostSimilar.Length);
}
public void testRescorer()
{
IDataModel dataModel = getDataModel(
new long[] { 1, 2, 3 },
new Double?[][] {
new double?[] { 0.1, 0.2 },
new double?[] { 0.2, 0.3, 0.3, 0.6 },
new double?[] { 0.4, 0.5, 0.5, 0.9 },
});
IUserSimilarity similarity = new PearsonCorrelationSimilarity(dataModel);
IUserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, dataModel);
IRecommender recommender = new GenericUserBasedRecommender(dataModel, neighborhood, similarity);
IList <IRecommendedItem> originalRecommended = recommender.Recommend(1, 2);
IList <IRecommendedItem> rescoredRecommended =
recommender.Recommend(1, 2, new ReversingRescorer <long>());
Assert.NotNull(originalRecommended);
Assert.NotNull(rescoredRecommended);
Assert.AreEqual(2, originalRecommended.Count);
Assert.AreEqual(2, rescoredRecommended.Count);
Assert.AreEqual(originalRecommended[0].GetItemID(), rescoredRecommended[1].GetItemID());
Assert.AreEqual(originalRecommended[1].GetItemID(), rescoredRecommended[0].GetItemID());
}
static void Main(string[] args)
{
// load movies
var movies = new List <MovieRecord>();
using (TextReader reader = File.OpenText(@"data/movies.csv"))
{
CsvReader csv = new CsvReader(reader);
csv.Configuration.Delimiter = ",";
csv.Configuration.MissingFieldFound = null;
while (csv.Read())
{
movies.Add(csv.GetRecord <MovieRecord>());
}
}
// load users
List <int> users = new List <int>();
using (TextReader reader = File.OpenText(@"data/ratings.csv"))
{
CsvReader csv = new CsvReader(reader);
csv.Configuration.Delimiter = ",";
csv.Configuration.MissingFieldFound = null;
while (csv.Read())
{
if (int.TryParse(csv.GetField(0), out int id))
{
if (!users.Contains(id))
{
users.Add(id);
}
}
}
}
// load data model
var model = new FileDataModel("data/ratings.csv", false, FileDataModel.DEFAULT_MIN_RELOAD_INTERVAL_MS, false);
//init values
string similarityType = "Cosine Similarity";
string recommenderType = "Userbased";
int neighborhoodSize = 125;
int resultCount = 10;
List <int> movieIdPredList = new List <int>();
List <int> userIdPredList = new List <int>();
//default recommender
AbstractSimilarity similarity = new UncenteredCosineSimilarity(model);
var neighborhood = new NearestNUserNeighborhood(neighborhoodSize, similarity, model);
AbstractRecommender recommender = new GenericUserBasedRecommender(model, neighborhood, similarity);
string input = "";
while (input != "X")
{
Console.WriteLine();
Console.WriteLine("Was möchten Sie tun?");
Console.WriteLine("I - Umschalten auf Item-based Ansatz");
Console.WriteLine("U - Umschalten auf User-based Ansatz");
Console.WriteLine("DE - Umschalten auf Cosine Similarity");
Console.WriteLine("PE - Umschalten auf Pearson Correlation");
Console.WriteLine("N - Anzahl der Nachbarn festlegen");
Console.WriteLine("C - Konfiguration neu erstellen und Datendatei neu lesen");
Console.WriteLine("O - Aktuelle Konfiguration ausgeben");
Console.WriteLine("E - Empfehlung durchführen");
Console.WriteLine("P - Prediction für spezifischen Film durchführen");
Console.WriteLine("UADD - Benutzer zur Liste für Vorschläge hinzufügen");
Console.WriteLine("MADD - Film zur Filmliste hinzufügen");
Console.WriteLine("UCLEAR - Benutzerliste leeren");
Console.WriteLine("MCLEAR - Filmliste leeren");
Console.WriteLine("X - Programm beenden");
input = Console.ReadLine();
if (input == "I")
{
recommenderType = "Itembased";
recommender = new GenericItemBasedRecommender(model, similarity);
Console.WriteLine("Recommendertyp auf \"itembased\" geändert.");
}
else if (input == "U")
{
recommenderType = "Userbased";
recommender = new GenericUserBasedRecommender(model, neighborhood, similarity);
Console.WriteLine("Recommendertyp auf \"userbased\" geändert.");
}
else if (input == "DE")
{
similarityType = "Cosine Similarity";
similarity = new UncenteredCosineSimilarity(model);
Console.WriteLine("Similaritytyp auf \"Cosine Similarity\" geändert.");
}
else if (input == "PE")
{
similarityType = "Pearson Correlation Similarity";
similarity = new PearsonCorrelationSimilarity(model);
Console.WriteLine("Similaritytyp auf \"Pearson Correlation Similarity\" geändert.");
}
else if (input == "N")
{
Console.WriteLine("Wie groß soll die Nachbarschaft sein?");
string neighborhoodSizeInput = Console.ReadLine();
if (int.TryParse(neighborhoodSizeInput, out int neighborhoodSizeParsed))
{
neighborhoodSize = neighborhoodSizeParsed;
neighborhood = new NearestNUserNeighborhood(neighborhoodSize, similarity, model);
Console.WriteLine("Nachbarschaftsgröße auf " + neighborhoodSize + " geändert.");
}
else
{
Console.WriteLine("Error: Es wurde keine Zahl eingegeben.");
Console.WriteLine("Die Nachbarschaftsgröße bleibt bei " + neighborhoodSize + ".");
}
}
else if (input == "C")
{
Console.WriteLine("Wie groß soll die Nachbarschaft sein?");
string neighborhoodSizeInput = Console.ReadLine();
if (int.TryParse(neighborhoodSizeInput, out int neighborhoodSizeParsed))
{
neighborhoodSize = neighborhoodSizeParsed;
neighborhood = new NearestNUserNeighborhood(neighborhoodSize, similarity, model);
Console.WriteLine("Nachbarschaftsgröße auf " + neighborhoodSize + " geändert.");
}
else
{
Console.WriteLine("Error: Es wurde keine Zahl eingegeben.");
Console.WriteLine("Die Nachbarschaftsgröße bleibt bei " + neighborhoodSize + ".");
}
Console.WriteLine("Wieviele Filme sollen vorgeschlagen werden?");
string resultCountInput = Console.ReadLine();
if (int.TryParse(resultCountInput, out int resultCountParsed))
{
resultCount = resultCountParsed;
}
else
{
Console.WriteLine("Error: Es wurde keine Zahl eingegeben.");
Console.WriteLine("Die Anzahl der vorgeschlagenen Filme bleibt bei " + resultCount + ".");
}
foreach (int userId in userIdPredList)
{
Console.WriteLine("Vorgeschlagene Filme für User mit UserId " + userId);
IList <IRecommendedItem> recommendedItems = recommender.Recommend(userId, resultCount);
foreach (var recItem in recommendedItems)
{
Console.WriteLine("Item: " + recItem.GetItemID() + " (" + movies.First(x => x.movieId == recItem.GetItemID()).title + ") ===> " + recItem.GetValue());
}
Console.WriteLine();
}
// load movies
movies = new List <MovieRecord>();
using (TextReader reader = File.OpenText(@"data/movies.csv"))
{
CsvReader csv = new CsvReader(reader);
csv.Configuration.Delimiter = ",";
csv.Configuration.MissingFieldFound = null;
while (csv.Read())
{
movies.Add(csv.GetRecord <MovieRecord>());
}
}
// load users
users = new List <int>();
using (TextReader reader = File.OpenText(@"data/ratings.csv"))
{
CsvReader csv = new CsvReader(reader);
csv.Configuration.Delimiter = ",";
csv.Configuration.MissingFieldFound = null;
while (csv.Read())
{
if (int.TryParse(csv.GetField(0), out int id))
{
if (!users.Contains(id))
{
users.Add(id);
}
}
}
}
model = new FileDataModel("data/ratings.csv", false, FileDataModel.DEFAULT_MIN_RELOAD_INTERVAL_MS, false);
if (similarityType == "Cosine Similarity")
{
similarity = new UncenteredCosineSimilarity(model);
}
else
{
similarity = new PearsonCorrelationSimilarity(model);
}
neighborhood = new NearestNUserNeighborhood(neighborhoodSize, similarity, model);
if (recommenderType == "Itembased")
{
recommender = new GenericItemBasedRecommender(model, similarity);
}
else
{
recommender = new GenericUserBasedRecommender(model, neighborhood, similarity);
}
}
else if (input == "O")
{
int countPredUser = userIdPredList.Count;
int countPredMovies = movieIdPredList.Count;
Console.WriteLine("-----------------------------------------------");
Console.WriteLine("Recommender Typ: " + recommenderType);
Console.WriteLine("Similarity Typ: " + similarityType);
Console.WriteLine("Nachbarschaftsgröße: " + neighborhoodSize);
Console.WriteLine("Anzahl Ergebnisse: " + resultCount);
Console.WriteLine("Anzahl Benutzer: " + countPredUser);
Console.WriteLine("Anzahl vorgeschlagene Filme: " + countPredMovies);
Console.WriteLine("-----------------------------------------------");
}
else if (input == "E")
{
Console.WriteLine("Wieviele Filme sollen vorgeschlagen werden?");
string resultCountInput = Console.ReadLine();
if (int.TryParse(resultCountInput, out int resultCountParsed))
{
resultCount = resultCountParsed;
}
else
{
Console.WriteLine("Error: Es wurde keine Zahl eingegeben.");
Console.WriteLine("Die Anzahl der vorgeschlagenen Filme bleibt bei " + resultCount + ".");
}
foreach (int userId in userIdPredList)
{
Console.WriteLine("Vorgeschlagene Filme für User mit UserId " + userId);
IList <IRecommendedItem> recommendedItems = recommender.Recommend(userId, resultCount);
foreach (var recItem in recommendedItems)
{
Console.WriteLine("Item: " + recItem.GetItemID() + " (" + movies.First(x => x.movieId == recItem.GetItemID()).title + ") ===> " + recItem.GetValue());
}
Console.WriteLine();
}
}
else if (input == "P")
{
foreach (int userId in userIdPredList)
{
Console.WriteLine("Prediction für User mit Id" + userId + ":");
foreach (int movieId in movieIdPredList)
{
float predictionValue = recommender.EstimatePreference(userId, movieId);
Console.WriteLine(movieId + " -> " + movies.First(x => x.movieId == movieId).title + " ===> " + predictionValue);
}
Console.WriteLine();
}
}
else if (input == "UADD")
{
Console.WriteLine("Welcher Benutzer (Id) soll zur Liste für Vorschläge hinzugefügt werden?");
string userIdInput = Console.ReadLine();
if (int.TryParse(userIdInput, out int userIdParsed))
{
if (users.Contains(userIdParsed))
{
userIdPredList.Add(userIdParsed);
}
else
{
Console.WriteLine("Error: UserId existiert nicht.");
}
}
else
{
Console.WriteLine("Error: Es wurde keine Zahl eingegeben.");
}
}
else if (input == "MADD")
{
Console.WriteLine("Welcher Film (Id) soll zur Liste für Vorschläge hinzugefügt werden?");
string movieIdInput = Console.ReadLine();
if (int.TryParse(movieIdInput, out int movieIdParsed))
{
bool movieIdFound = false;
foreach (MovieRecord movie in movies)
{
if (movie.movieId == movieIdParsed)
{
movieIdPredList.Add(movieIdParsed);
movieIdFound = true;
break;
}
}
if (!movieIdFound)
{
Console.WriteLine("Error: movieId existiert nicht. Es wurde kein Film zur Liste hinzugefügt");
}
}
else
{
Console.WriteLine("Error: Es wurde keine Zahl eingegeben.");
}
}
else if (input == "UCLEAR")
{
userIdPredList.Clear();
}
else if (input == "MCLEAR")
{
movieIdPredList.Clear();
}
}
Console.WriteLine("Programm beendet");
Console.ReadLine();
}