private static List<int[]> getPopularResources(BookmarkReader reader, int count, int trainSize) {
List<int[]> resources = new ArrayList<int[]>();
Map<Integer, Integer> countMap = new LinkedHashMap<Integer, Integer>();
for (int i = 0; i < reader.getResources().size(); i++) {
countMap.put(i, reader.getResourceCounts().get(i));
}
Map<Integer, Integer> sortedCountMap =
new TreeMap<Integer, Integer>(new IntMapComparator(countMap));
sortedCountMap.putAll(countMap);
for (int userID : reader.getUniqueUserListFromTestSet(trainSize)) {
List<Integer> userResources =
UserData.getResourcesFromUser(reader.getUserLines().subList(0, trainSize), userID);
// System.out.println(userResources.size());
List<Integer> resIDs = new ArrayList<Integer>();
int i = 0;
for (Integer key : sortedCountMap.keySet()) {
if (i < count) {
if (!userResources.contains(key)) {
resIDs.add(key);
i++;
}
} else {
break;
}
}
resources.add(Ints.toArray(resIDs));
}
return resources;
}
private static void evaluateAllTagRecommenderApproaches(String sampleDir, String samplePath) {
getTrainTestSize(samplePath);
BookmarkReader reader = new BookmarkReader(TRAIN_SIZE, false);
reader.readFile(samplePath);
evaluate(sampleDir, samplePath, "apr", null, true, false, reader);
evaluate(sampleDir, samplePath, "bll_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "bll_ac_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "bll_c_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "bll_c_ac_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "cf_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "fr", null, true, false, reader);
evaluate(sampleDir, samplePath, "girp", null, true, false, reader);
evaluate(sampleDir, samplePath, "girptm", null, true, false, reader);
evaluate(sampleDir, samplePath, "layers_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "layerstagbll_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "layerstopicbll_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "lda_1000", null, true, false, reader);
evaluate(sampleDir, samplePath, "mp", null, true, false, reader);
evaluate(sampleDir, samplePath, "mp_r_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "mp_u_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "mp_ur_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "rescf_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "usercf_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "userlayers_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "userlayerstagbll_5_5", null, true, false, reader);
evaluate(sampleDir, samplePath, "userlayerstopicbll_5_5", null, true, false, reader);
}
public static void predictRandomResources(String filename, int trainSize) {
// filename += "_res";
// TODO: do not use complete size
BookmarkReader reader = new BookmarkReader(0, false);
reader.readFile(filename);
List<int[]> values = getRandomResources(reader, 10, trainSize);
PredictionFileWriter writer = new PredictionFileWriter(reader, values);
writer.writeResourcePredictionsToFile(filename + "_rand", trainSize, 0);
}
private static List<int[]> getPerfectTags(BookmarkReader reader, int sampleSize, int limit) {
List<int[]> tags = new ArrayList<int[]>();
int trainSize = reader.getUserLines().size() - sampleSize;
for (UserData data : reader.getUserLines().subList(trainSize, trainSize + sampleSize)) {
List<Integer> t = data.getTags();
while (t.size() < limit) {
t.add(-1);
}
tags.add(Ints.toArray(t));
}
return tags;
}
public static void predictPopularTags(String filename, int trainSize, int sampleSize) {
// filename += "_res";
BookmarkReader reader = new BookmarkReader(trainSize, false);
reader.readFile(filename);
List<int[]> values = getPopularTags(reader, sampleSize, 10);
// List<int[]> values = getPerfectTags(reader, sampleSize, 10);
reader.setUserLines(reader.getUserLines().subList(trainSize, reader.getUserLines().size()));
PredictionFileWriter writer = new PredictionFileWriter(reader, values);
writer.writeFile(filename + "_mp");
Utilities.writeStringToFile("./data/metrics/" + filename + "_mp" + "_TIME.txt", timeString);
}
public static int[] getPopularTagList(BookmarkReader reader, int size) {
Map<Integer, Integer> countMap = new LinkedHashMap<Integer, Integer>();
for (int i = 0; i < reader.getTagCounts().size(); i++) {
countMap.put(i, reader.getTagCounts().get(i));
}
Map<Integer, Integer> sortedCountMap =
new TreeMap<Integer, Integer>(new IntMapComparator(countMap));
sortedCountMap.putAll(countMap);
int[] tagIDs = new int[size];
int i = 0;
for (Integer key : sortedCountMap.keySet()) {
if (i < size) {
tagIDs[i++] = key;
} else {
break;
}
}
return tagIDs;
}
private static List<int[]> getRandomResources(BookmarkReader reader, int count, int trainSize) {
List<int[]> resources = new ArrayList<int[]>();
int resCount = reader.getResources().size();
for (int userID : reader.getUniqueUserListFromTestSet(trainSize)) {
List<Integer> userResources =
UserData.getResourcesFromUser(reader.getUserLines().subList(0, trainSize), userID);
List<Integer> resIDs = new ArrayList<Integer>();
int i = 0;
for (Integer res : Utilities.getRandomIndices(0, resCount - 1)) {
if (i < count) {
if (!userResources.contains(res)) {
resIDs.add(res);
i++;
}
} else {
break;
}
}
resources.add(Ints.toArray(resIDs));
}
return resources;
}
private static void getTrainTestSize(String sample) {
if (TOPIC_NAME != null) {
sample += ("_" + TOPIC_NAME);
}
BookmarkReader trainReader = new BookmarkReader(-1, false);
trainReader.readFile(sample + "_train");
TRAIN_SIZE = trainReader.getBookmarks().size();
System.out.println("Train-size: " + TRAIN_SIZE);
BookmarkReader testReader = new BookmarkReader(-1, false);
testReader.readFile(sample + "_test");
TEST_SIZE = testReader.getBookmarks().size();
System.out.println("Test-size: " + TEST_SIZE);
}
// passing the trainSize means that MyMediaLite files will be evaluated
private static void evaluate(
String sampleDir,
String sampleName,
String prefix,
String postfix,
boolean calcTags,
boolean tensor,
BookmarkReader reader) {
if (reader == null) {
getTrainTestSize(sampleName + (postfix != null ? "_" + postfix : ""));
reader = new BookmarkReader(TRAIN_SIZE, false);
reader.readFile(sampleName + (postfix != null ? "_" + postfix : ""));
}
if (calcTags) {
writeMetrics(
sampleDir, sampleName, prefix, 1, 10, postfix, reader, tensor ? TRAIN_SIZE : null);
} else {
writeMetricsForResources(
sampleDir, sampleName, prefix, 1, 20, postfix, reader, tensor ? TRAIN_SIZE : null);
}
}
private static void getStatistics(String dataset, boolean writeAll) {
if (TOPIC_NAME != null) {
dataset += ("_" + TOPIC_NAME);
}
BookmarkReader reader = new BookmarkReader(0, false);
reader.readFile(dataset);
int bookmarks = reader.getBookmarks().size();
System.out.println("Posts: " + bookmarks);
int users = reader.getUsers().size();
System.out.println("Users: " + users);
int resources = reader.getResources().size();
System.out.println("Resources: " + resources);
int tags = reader.getTags().size();
System.out.println("Tags: " + tags);
int tagAssignments = reader.getTagAssignmentsCount();
System.out.println("Tag-Assignments: " + tagAssignments);
int categories = reader.getCategories().size();
System.out.println("Topics: " + categories);
double avgTASPerPost = (double) tagAssignments / bookmarks;
System.out.println("Avg. TAS per post: " + avgTASPerPost);
double avgBookmarksPerUser = (double) bookmarks / users;
System.out.println("Avg. resources/posts per user: "******"Avg. users/posts per resource: " + avgBookmarksPerResource);
if (writeAll) {
try {
getTrainTestSize(dataset);
FileWriter userWriter =
new FileWriter(new File("./data/metrics/" + dataset + "_userStats.txt"));
BufferedWriter userBW = new BufferedWriter(userWriter);
userBW.write("UserID| NoOfResources| NoOfTopics| Topic-Similarity\n");
List<Bookmark> trainList = reader.getBookmarks().subList(0, TRAIN_SIZE);
List<Integer> testUsers = reader.getUniqueUserListFromTestSet(TRAIN_SIZE);
System.out.println();
double avgTopicsPerUser = 0.0;
double avgTopicDiversityPerUser = 0.0;
List<Map<Integer, Double>> userTopics = Utilities.getRelativeTopicMaps(trainList, false);
List<List<Bookmark>> userBookmarks = Utilities.getBookmarks(trainList, false);
for (int userID : testUsers) {
Map<Integer, Double> topicsOfUser = userTopics.get(userID);
double topicDiversityOfUser = Bookmark.getBookmarkDiversity(userBookmarks.get(userID));
userBW.write(
userID
+ "| "
+ reader.getUserCounts().get(userID)
+ "| "
+ topicsOfUser.keySet().size()
+ "| "
+ topicDiversityOfUser
+ "\n");
avgTopicsPerUser += topicsOfUser.keySet().size();
avgTopicDiversityPerUser += topicDiversityOfUser;
}
System.out.println("Avg. topics per user: "******"Avg. topic-similarity per user: "******"Avg. topics per resource: " + avgTopicsPerResource);
userBW.flush();
userBW.close();
} catch (IOException e) {
System.out.println(e.getMessage());
}
}
System.out.println();
}