Merge pull request #7503 from alimate/BAEL-3070

BAEL-3070: K-Means Clustering Code Samples

algorithms-miscellaneous-3/pom.xml

@@ -31,6 +31,17 @@
             <version>${guava.version}</version>
         </dependency>
 
+        <dependency>
+            <groupId>com.squareup.retrofit2</groupId>
+            <artifactId>retrofit</artifactId>
+            <version>${retrofit.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>com.squareup.retrofit2</groupId>
+            <artifactId>converter-jackson</artifactId>
+            <version>${retrofit.version}</version>
+        </dependency>
+
         <dependency>
             <groupId>org.apache.commons</groupId>
             <artifactId>commons-lang3</artifactId>
@@ -61,5 +72,6 @@
         <org.assertj.core.version>3.9.0</org.assertj.core.version>
         <commons-collections4.version>4.3</commons-collections4.version>
         <guava.version>28.0-jre</guava.version>
+        <retrofit.version>2.6.0</retrofit.version>
     </properties>
 </project>

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/Centroid.java

@@ -0,0 +1,45 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.util.Map;
+import java.util.Objects;
+
+/**
+ * Encapsulates all coordinates for a particular cluster centroid.
+ */
+public class Centroid {
+
+    /**
+     * The centroid coordinates.
+     */
+    private final Map<String, Double> coordinates;
+
+    public Centroid(Map<String, Double> coordinates) {
+        this.coordinates = coordinates;
+    }
+
+    public Map<String, Double> getCoordinates() {
+        return coordinates;
+    }
+
+    @Override
+    public boolean equals(Object o) {
+        if (this == o) {
+            return true;
+        }
+        if (o == null || getClass() != o.getClass()) {
+            return false;
+        }
+        Centroid centroid = (Centroid) o;
+        return Objects.equals(getCoordinates(), centroid.getCoordinates());
+    }
+
+    @Override
+    public int hashCode() {
+        return Objects.hash(getCoordinates());
+    }
+
+    @Override
+    public String toString() {
+        return "Centroid " + coordinates;
+    }
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/Distance.java

@@ -0,0 +1,20 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.util.Map;
+
+/**
+ * Defines a contract to calculate distance between two feature vectors. The less the
+ * calculated distance, the more two items are similar to each other.
+ */
+public interface Distance {
+
+    /**
+     * Calculates the distance between two feature vectors.
+     *
+     * @param f1 The first set of features.
+     * @param f2 The second set of features.
+     * @return Calculated distance.
+     * @throws IllegalArgumentException If the given feature vectors are invalid.
+     */
+    double calculate(Map<String, Double> f1, Map<String, Double> f2);
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/Errors.java

@@ -0,0 +1,23 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Encapsulates methods to calculates errors between centroid and the cluster members.
+ */
+public class Errors {
+
+    public static double sse(Map<Centroid, List<Record>> clustered, Distance distance) {
+        double sum = 0;
+        for (Map.Entry<Centroid, List<Record>> entry : clustered.entrySet()) {
+            Centroid centroid = entry.getKey();
+            for (Record record : entry.getValue()) {
+                double d = distance.calculate(centroid.getCoordinates(), record.getFeatures());
+                sum += Math.pow(d, 2);
+            }
+        }
+
+        return sum;
+    }
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/EuclideanDistance.java

@@ -0,0 +1,26 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.util.Map;
+
+/**
+ * Calculates the distance between two items using the Euclidean formula.
+ */
+public class EuclideanDistance implements Distance {
+
+    @Override
+    public double calculate(Map<String, Double> f1, Map<String, Double> f2) {
+        if (f1 == null || f2 == null) {
+            throw new IllegalArgumentException("Feature vectors can't be null");
+        }
+
+        double sum = 0;
+        for (String key : f1.keySet()) {
+            Double v1 = f1.get(key);
+            Double v2 = f2.get(key);
+
+            if (v1 != null && v2 != null) sum += Math.pow(v1 - v2, 2);
+        }
+
+        return Math.sqrt(sum);
+    }
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/KMeans.java

@@ -0,0 +1,236 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Random;
+import java.util.Set;
+
+import static java.util.stream.Collectors.toList;
+import static java.util.stream.Collectors.toSet;
+
+/**
+ * Encapsulates an implementation of KMeans clustering algorithm.
+ *
+ * @author Ali Dehghani
+ */
+public class KMeans {
+
+    private KMeans() {
+        throw new IllegalAccessError("You shouldn't call this constructor");
+    }
+
+    /**
+     * Will be used to generate random numbers.
+     */
+    private static final Random random = new Random();
+
+    /**
+     * Performs the K-Means clustering algorithm on the given dataset.
+     *
+     * @param records       The dataset.
+     * @param k             Number of Clusters.
+     * @param distance      To calculate the distance between two items.
+     * @param maxIterations Upper bound for the number of iterations.
+     * @return K clusters along with their features.
+     */
+    public static Map<Centroid, List<Record>> fit(List<Record> records, int k, Distance distance, int maxIterations) {
+        applyPreconditions(records, k, distance, maxIterations);
+
+        List<Centroid> centroids = randomCentroids(records, k);
+        Map<Centroid, List<Record>> clusters = new HashMap<>();
+        Map<Centroid, List<Record>> lastState = new HashMap<>();
+
+        // iterate for a pre-defined number of times
+        for (int i = 0; i < maxIterations; i++) {
+            boolean isLastIteration = i == maxIterations - 1;
+
+            // in each iteration we should find the nearest centroid for each record
+            for (Record record : records) {
+                Centroid centroid = nearestCentroid(record, centroids, distance);
+                assignToCluster(clusters, record, centroid);
+            }
+
+            // if the assignment does not change, then the algorithm terminates
+            boolean shouldTerminate = isLastIteration || clusters.equals(lastState);
+            lastState = clusters;
+            if (shouldTerminate) {
+                break;
+            }
+
+            // at the end of each iteration we should relocate the centroids
+            centroids = relocateCentroids(clusters);
+            clusters = new HashMap<>();
+        }
+
+        return lastState;
+    }
+
+    /**
+     * Move all cluster centroids to the average of all assigned features.
+     *
+     * @param clusters The current cluster configuration.
+     * @return Collection of new and relocated centroids.
+     */
+    private static List<Centroid> relocateCentroids(Map<Centroid, List<Record>> clusters) {
+        return clusters
+          .entrySet()
+          .stream()
+          .map(e -> average(e.getKey(), e.getValue()))
+          .collect(toList());
+    }
+
+    /**
+     * Moves the given centroid to the average position of all assigned features. If
+     * the centroid has no feature in its cluster, then there would be no need for a
+     * relocation. Otherwise, for each entry we calculate the average of all records
+     * first by summing all the entries and then dividing the final summation value by
+     * the number of records.
+     *
+     * @param centroid The centroid to move.
+     * @param records  The assigned features.
+     * @return The moved centroid.
+     */
+    private static Centroid average(Centroid centroid, List<Record> records) {
+        // if this cluster is empty, then we shouldn't move the centroid
+        if (records == null || records.isEmpty()) {
+            return centroid;
+        }
+
+        // Since some records don't have all possible attributes, we initialize
+        // average coordinates equal to current centroid coordinates
+        Map<String, Double> average = centroid.getCoordinates();
+
+        // The average function works correctly if we clear all coordinates corresponding
+        // to present record attributes
+        records
+          .stream()
+          .flatMap(e -> e
+            .getFeatures()
+            .keySet()
+            .stream())
+          .forEach(k -> average.put(k, 0.0));
+
+        for (Record record : records) {
+            record
+              .getFeatures()
+              .forEach((k, v) -> average.compute(k, (k1, currentValue) -> v + currentValue));
+        }
+
+        average.forEach((k, v) -> average.put(k, v / records.size()));
+
+        return new Centroid(average);
+    }
+
+    /**
+     * Assigns a feature vector to the given centroid. If this is the first assignment for this centroid,
+     * first we should create the list.
+     *
+     * @param clusters The current cluster configuration.
+     * @param record   The feature vector.
+     * @param centroid The centroid.
+     */
+    private static void assignToCluster(Map<Centroid, List<Record>> clusters, Record record, Centroid centroid) {
+        clusters.compute(centroid, (key, list) -> {
+            if (list == null) {
+                list = new ArrayList<>();
+            }
+
+            list.add(record);
+            return list;
+        });
+    }
+
+    /**
+     * With the help of the given distance calculator, iterates through centroids and finds the
+     * nearest one to the given record.
+     *
+     * @param record    The feature vector to find a centroid for.
+     * @param centroids Collection of all centroids.
+     * @param distance  To calculate the distance between two items.
+     * @return The nearest centroid to the given feature vector.
+     */
+    private static Centroid nearestCentroid(Record record, List<Centroid> centroids, Distance distance) {
+        double minimumDistance = Double.MAX_VALUE;
+        Centroid nearest = null;
+
+        for (Centroid centroid : centroids) {
+            double currentDistance = distance.calculate(record.getFeatures(), centroid.getCoordinates());
+
+            if (currentDistance < minimumDistance) {
+                minimumDistance = currentDistance;
+                nearest = centroid;
+            }
+        }
+
+        return nearest;
+    }
+
+    /**
+     * Generates k random centroids. Before kicking-off the centroid generation process,
+     * first we calculate the possible value range for each attribute. Then when
+     * we're going to generate the centroids, we generate random coordinates in
+     * the [min, max] range for each attribute.
+     *
+     * @param records The dataset which helps to calculate the [min, max] range for
+     *                each attribute.
+     * @param k       Number of clusters.
+     * @return Collections of randomly generated centroids.
+     */
+    private static List<Centroid> randomCentroids(List<Record> records, int k) {
+        List<Centroid> centroids = new ArrayList<>();
+        Map<String, Double> maxs = new HashMap<>();
+        Map<String, Double> mins = new HashMap<>();
+
+        for (Record record : records) {
+            record
+              .getFeatures()
+              .forEach((key, value) -> {
+                  // compares the value with the current max and choose the bigger value between them
+                  maxs.compute(key, (k1, max) -> max == null || value > max ? value : max);
+
+                  // compare the value with the current min and choose the smaller value between them
+                  mins.compute(key, (k1, min) -> min == null || value < min ? value : min);
+              });
+        }
+
+        Set<String> attributes = records
+          .stream()
+          .flatMap(e -> e
+            .getFeatures()
+            .keySet()
+            .stream())
+          .collect(toSet());
+        for (int i = 0; i < k; i++) {
+            Map<String, Double> coordinates = new HashMap<>();
+            for (String attribute : attributes) {
+                double max = maxs.get(attribute);
+                double min = mins.get(attribute);
+                coordinates.put(attribute, random.nextDouble() * (max - min) + min);
+            }
+
+            centroids.add(new Centroid(coordinates));
+        }
+
+        return centroids;
+    }
+
+    private static void applyPreconditions(List<Record> records, int k, Distance distance, int maxIterations) {
+        if (records == null || records.isEmpty()) {
+            throw new IllegalArgumentException("The dataset can't be empty");
+        }
+
+        if (k <= 1) {
+            throw new IllegalArgumentException("It doesn't make sense to have less than or equal to 1 cluster");
+        }
+
+        if (distance == null) {
+            throw new IllegalArgumentException("The distance calculator is required");
+        }
+
+        if (maxIterations <= 0) {
+            throw new IllegalArgumentException("Max iterations should be a positive number");
+        }
+    }
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/LastFm.java

@@ -0,0 +1,144 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import com.fasterxml.jackson.databind.ObjectMapper;
+import okhttp3.OkHttpClient;
+import retrofit2.Retrofit;
+import retrofit2.converter.jackson.JacksonConverterFactory;
+
+import static java.util.stream.Collectors.toSet;
+
+public class LastFm {
+
+    private static OkHttpClient okHttp = new OkHttpClient.Builder()
+      .addInterceptor(new LastFmService.Authenticator("put your API key here"))
+      .build();
+
+    private static Retrofit retrofit = new Retrofit.Builder()
+      .client(okHttp)
+      .addConverterFactory(JacksonConverterFactory.create())
+      .baseUrl("http://ws.audioscrobbler.com/")
+      .build();
+
+    private static LastFmService lastFm = retrofit.create(LastFmService.class);
+
+    private static ObjectMapper mapper = new ObjectMapper();
+
+    public static void main(String[] args) throws IOException {
+        List<String> artists = getTop100Artists();
+        Set<String> tags = getTop100Tags();
+        List<Record> records = datasetWithTaggedArtists(artists, tags);
+
+        Map<Centroid, List<Record>> clusters = KMeans.fit(records, 7, new EuclideanDistance(), 1000);
+        // Print the cluster configuration
+        clusters.forEach((key, value) -> {
+            System.out.println("------------------------------ CLUSTER -----------------------------------");
+
+            System.out.println(sortedCentroid(key));
+            String members = String.join(", ", value
+              .stream()
+              .map(Record::getDescription)
+              .collect(toSet()));
+            System.out.print(members);
+
+            System.out.println();
+            System.out.println();
+        });
+
+        Map<String, Object> json = convertToD3CompatibleMap(clusters);
+        System.out.println(mapper.writeValueAsString(json));
+    }
+
+    private static Map<String, Object> convertToD3CompatibleMap(Map<Centroid, List<Record>> clusters) {
+        Map<String, Object> json = new HashMap<>();
+        json.put("name", "Musicians");
+        List<Map<String, Object>> children = new ArrayList<>();
+        clusters.forEach((key, value) -> {
+            Map<String, Object> child = new HashMap<>();
+            child.put("name", dominantGenre(sortedCentroid(key)));
+            List<Map<String, String>> nested = new ArrayList<>();
+            for (Record record : value) {
+                nested.add(Collections.singletonMap("name", record.getDescription()));
+            }
+            child.put("children", nested);
+
+            children.add(child);
+        });
+        json.put("children", children);
+        return json;
+    }
+
+    private static String dominantGenre(Centroid centroid) {
+        return centroid
+          .getCoordinates()
+          .keySet()
+          .stream()
+          .limit(2)
+          .collect(Collectors.joining(", "));
+    }
+
+    private static Centroid sortedCentroid(Centroid key) {
+        List<Map.Entry<String, Double>> entries = new ArrayList<>(key
+          .getCoordinates()
+          .entrySet());
+        entries.sort((e1, e2) -> e2
+          .getValue()
+          .compareTo(e1.getValue()));
+
+        Map<String, Double> sorted = new LinkedHashMap<>();
+        for (Map.Entry<String, Double> entry : entries) {
+            sorted.put(entry.getKey(), entry.getValue());
+        }
+
+        return new Centroid(sorted);
+    }
+
+    private static List<Record> datasetWithTaggedArtists(List<String> artists, Set<String> topTags) throws IOException {
+        List<Record> records = new ArrayList<>();
+        for (String artist : artists) {
+            Map<String, Double> tags = lastFm
+              .topTagsFor(artist)
+              .execute()
+              .body()
+              .all();
+
+            // Only keep popular tags.
+            tags
+              .entrySet()
+              .removeIf(e -> !topTags.contains(e.getKey()));
+
+            records.add(new Record(artist, tags));
+        }
+        return records;
+    }
+
+    private static Set<String> getTop100Tags() throws IOException {
+        return lastFm
+          .topTags()
+          .execute()
+          .body()
+          .all();
+    }
+
+    private static List<String> getTop100Artists() throws IOException {
+        List<String> artists = new ArrayList<>();
+        for (int i = 1; i <= 2; i++) {
+            artists.addAll(lastFm
+              .topArtists(i)
+              .execute()
+              .body()
+              .all());
+        }
+
+        return artists;
+    }
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/LastFmService.java

@@ -0,0 +1,118 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import com.fasterxml.jackson.annotation.JsonAutoDetect;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import okhttp3.HttpUrl;
+import okhttp3.Interceptor;
+import okhttp3.Request;
+import okhttp3.Response;
+import retrofit2.Call;
+import retrofit2.http.GET;
+import retrofit2.http.Query;
+
+import static com.fasterxml.jackson.annotation.JsonAutoDetect.Visibility.ANY;
+import static java.util.stream.Collectors.toList;
+
+public interface LastFmService {
+
+    @GET("/2.0/?method=chart.gettopartists&format=json&limit=50")
+    Call<Artists> topArtists(@Query("page") int page);
+
+    @GET("/2.0/?method=artist.gettoptags&format=json&limit=20&autocorrect=1")
+    Call<Tags> topTagsFor(@Query("artist") String artist);
+
+    @GET("/2.0/?method=chart.gettoptags&format=json&limit=100")
+    Call<TopTags> topTags();
+
+    /**
+     * HTTP interceptor to intercept all HTTP requests and add the API key to them.
+     */
+    class Authenticator implements Interceptor {
+
+        private final String apiKey;
+
+        Authenticator(String apiKey) {
+            this.apiKey = apiKey;
+        }
+
+        @Override
+        public Response intercept(Chain chain) throws IOException {
+            HttpUrl url = chain
+              .request()
+              .url()
+              .newBuilder()
+              .addQueryParameter("api_key", apiKey)
+              .build();
+            Request request = chain
+              .request()
+              .newBuilder()
+              .url(url)
+              .build();
+
+            return chain.proceed(request);
+        }
+    }
+
+    @JsonAutoDetect(fieldVisibility = ANY)
+    class TopTags {
+
+        private Map<String, Object> tags;
+
+        @SuppressWarnings("unchecked")
+        public Set<String> all() {
+            List<Map<String, Object>> topTags = (List<Map<String, Object>>) tags.get("tag");
+            return topTags
+              .stream()
+              .map(e -> ((String) e.get("name")))
+              .collect(Collectors.toSet());
+        }
+    }
+
+    @JsonAutoDetect(fieldVisibility = ANY)
+    class Tags {
+
+        @JsonProperty("toptags") private Map<String, Object> topTags;
+
+        @SuppressWarnings("unchecked")
+        public Map<String, Double> all() {
+            try {
+                Map<String, Double> all = new HashMap<>();
+                List<Map<String, Object>> tags = (List<Map<String, Object>>) topTags.get("tag");
+                for (Map<String, Object> tag : tags) {
+                    all.put(((String) tag.get("name")), ((Integer) tag.get("count")).doubleValue());
+                }
+
+                return all;
+            } catch (Exception e) {
+                return Collections.emptyMap();
+            }
+        }
+    }
+
+    @JsonAutoDetect(fieldVisibility = ANY)
+    class Artists {
+
+        private Map<String, Object> artists;
+
+        @SuppressWarnings("unchecked")
+        public List<String> all() {
+            try {
+                List<Map<String, Object>> artists = (List<Map<String, Object>>) this.artists.get("artist");
+                return artists
+                  .stream()
+                  .map(e -> ((String) e.get("name")))
+                  .collect(toList());
+            } catch (Exception e) {
+                return Collections.emptyList();
+            }
+        }
+    }
+}

algorithms-miscellaneous-3/src/main/java/com/baeldung/algorithms/kmeans/Record.java

@@ -0,0 +1,65 @@
+package com.baeldung.algorithms.kmeans;
+
+import java.util.Map;
+import java.util.Objects;
+
+/**
+ * Encapsulates all feature values for a few attributes. Optionally each record
+ * can be described with the {@link #description} field.
+ */
+public class Record {
+
+    /**
+     * The record description. For example, this can be the artist name for the famous musician
+     * example.
+     */
+    private final String description;
+
+    /**
+     * Encapsulates all attributes and their corresponding values, i.e. features.
+     */
+    private final Map<String, Double> features;
+
+    public Record(String description, Map<String, Double> features) {
+        this.description = description;
+        this.features = features;
+    }
+
+    public Record(Map<String, Double> features) {
+        this("", features);
+    }
+
+    public String getDescription() {
+        return description;
+    }
+
+    public Map<String, Double> getFeatures() {
+        return features;
+    }
+
+    @Override
+    public String toString() {
+        String prefix = description == null || description
+          .trim()
+          .isEmpty() ? "Record" : description;
+
+        return prefix + ": " + features;
+    }
+
+    @Override
+    public boolean equals(Object o) {
+        if (this == o) {
+            return true;
+        }
+        if (o == null || getClass() != o.getClass()) {
+            return false;
+        }
+        Record record = (Record) o;
+        return Objects.equals(getDescription(), record.getDescription()) && Objects.equals(getFeatures(), record.getFeatures());
+    }
+
+    @Override
+    public int hashCode() {
+        return Objects.hash(getDescription(), getFeatures());
+    }
+}

algorithms-miscellaneous-3/src/main/resources/kmeans/lastfm.json

@@ -0,0 +1,490 @@
+{
+  "children": [
+    {
+      "children": [
+        {
+          "name": "Radiohead"
+        },
+        {
+          "name": "Red Hot Chili Peppers"
+        },
+        {
+          "name": "Coldplay"
+        },
+        {
+          "name": "Nirvana"
+        },
+        {
+          "name": "Panic! at the Disco"
+        },
+        {
+          "name": "The Cure"
+        },
+        {
+          "name": "Linkin Park"
+        },
+        {
+          "name": "Radiohead"
+        },
+        {
+          "name": "Red Hot Chili Peppers"
+        },
+        {
+          "name": "Coldplay"
+        },
+        {
+          "name": "Nirvana"
+        },
+        {
+          "name": "Panic! at the Disco"
+        },
+        {
+          "name": "The Cure"
+        },
+        {
+          "name": "Linkin Park"
+        },
+        {
+          "name": "Muse"
+        },
+        {
+          "name": "Maroon 5"
+        },
+        {
+          "name": "Foo Fighters"
+        },
+        {
+          "name": "Paramore"
+        },
+        {
+          "name": "Oasis"
+        },
+        {
+          "name": "Fall Out Boy"
+        },
+        {
+          "name": "OneRepublic"
+        },
+        {
+          "name": "Weezer"
+        },
+        {
+          "name": "System of a Down"
+        },
+        {
+          "name": "The White Stripes"
+        }
+      ],
+      "name": "rock, alternative"
+    },
+    {
+      "children": [
+        {
+          "name": "Lil Nas X"
+        },
+        {
+          "name": "Post Malone"
+        },
+        {
+          "name": "Drake"
+        },
+        {
+          "name": "Kanye West"
+        },
+        {
+          "name": "Kendrick Lamar"
+        },
+        {
+          "name": "Tyler, the Creator"
+        },
+        {
+          "name": "Eminem"
+        },
+        {
+          "name": "Childish Gambino"
+        },
+        {
+          "name": "Frank Ocean"
+        },
+        {
+          "name": "Lil Nas X"
+        },
+        {
+          "name": "Post Malone"
+        },
+        {
+          "name": "Drake"
+        },
+        {
+          "name": "Kanye West"
+        },
+        {
+          "name": "Kendrick Lamar"
+        },
+        {
+          "name": "Tyler, the Creator"
+        },
+        {
+          "name": "Eminem"
+        },
+        {
+          "name": "Childish Gambino"
+        },
+        {
+          "name": "Frank Ocean"
+        },
+        {
+          "name": "Lizzo"
+        },
+        {
+          "name": "Travi$ Scott"
+        },
+        {
+          "name": "A$AP Rocky"
+        },
+        {
+          "name": "Nicki Minaj"
+        },
+        {
+          "name": "xxxtentacion"
+        }
+      ],
+      "name": "Hip-Hop, rap"
+    },
+    {
+      "children": [
+        {
+          "name": "Arctic Monkeys"
+        },
+        {
+          "name": "Imagine Dragons"
+        },
+        {
+          "name": "The Killers"
+        },
+        {
+          "name": "Gorillaz"
+        },
+        {
+          "name": "The Black Keys"
+        },
+        {
+          "name": "Arctic Monkeys"
+        },
+        {
+          "name": "Imagine Dragons"
+        },
+        {
+          "name": "The Killers"
+        },
+        {
+          "name": "Gorillaz"
+        },
+        {
+          "name": "The Black Keys"
+        },
+        {
+          "name": "Twenty One Pilots"
+        },
+        {
+          "name": "Ellie Goulding"
+        },
+        {
+          "name": "Florence + the Machine"
+        },
+        {
+          "name": "Vampire Weekend"
+        },
+        {
+          "name": "The Smiths"
+        },
+        {
+          "name": "The Strokes"
+        },
+        {
+          "name": "MGMT"
+        },
+        {
+          "name": "Foster the People"
+        },
+        {
+          "name": "Two Door Cinema Club"
+        },
+        {
+          "name": "Cage the Elephant"
+        },
+        {
+          "name": "Arcade Fire"
+        },
+        {
+          "name": "The 1975"
+        }
+      ],
+      "name": "indie, alternative"
+    },
+    {
+      "children": [
+        {
+          "name": "Ed Sheeran"
+        },
+        {
+          "name": "Tame Impala"
+        },
+        {
+          "name": "Ed Sheeran"
+        },
+        {
+          "name": "Tame Impala"
+        },
+        {
+          "name": "Green Day"
+        },
+        {
+          "name": "Metallica"
+        },
+        {
+          "name": "blink-182"
+        },
+        {
+          "name": "Bon Iver"
+        },
+        {
+          "name": "The Clash"
+        }
+      ],
+      "name": "rock, punk rock"
+    },
+    {
+      "children": [
+        {
+          "name": "Calvin Harris"
+        },
+        {
+          "name": "The Weeknd"
+        },
+        {
+          "name": "The Chainsmokers"
+        },
+        {
+          "name": "Daft Punk"
+        },
+        {
+          "name": "Marshmello"
+        },
+        {
+          "name": "David Guetta"
+        },
+        {
+          "name": "Calvin Harris"
+        },
+        {
+          "name": "The Weeknd"
+        },
+        {
+          "name": "The Chainsmokers"
+        },
+        {
+          "name": "Daft Punk"
+        },
+        {
+          "name": "Marshmello"
+        },
+        {
+          "name": "David Guetta"
+        },
+        {
+          "name": "Avicii"
+        },
+        {
+          "name": "Kygo"
+        },
+        {
+          "name": "Martin Garrix"
+        },
+        {
+          "name": "Major Lazer"
+        },
+        {
+          "name": "Depeche Mode"
+        }
+      ],
+      "name": "electronic, dance"
+    },
+    {
+      "children": [
+        {
+          "name": "Queen"
+        },
+        {
+          "name": "The Beatles"
+        },
+        {
+          "name": "David Bowie"
+        },
+        {
+          "name": "Fleetwood Mac"
+        },
+        {
+          "name": "Pink Floyd"
+        },
+        {
+          "name": "The Rolling Stones"
+        },
+        {
+          "name": "Led Zeppelin"
+        },
+        {
+          "name": "Queen"
+        },
+        {
+          "name": "The Beatles"
+        },
+        {
+          "name": "David Bowie"
+        },
+        {
+          "name": "Fleetwood Mac"
+        },
+        {
+          "name": "Pink Floyd"
+        },
+        {
+          "name": "The Rolling Stones"
+        },
+        {
+          "name": "Led Zeppelin"
+        },
+        {
+          "name": "Elton John"
+        }
+      ],
+      "name": "classic rock, rock"
+    },
+    {
+      "children": [
+        {
+          "name": "Billie Eilish"
+        },
+        {
+          "name": "Ariana Grande"
+        },
+        {
+          "name": "Taylor Swift"
+        },
+        {
+          "name": "Beyoncé"
+        },
+        {
+          "name": "Shawn Mendes"
+        },
+        {
+          "name": "Rihanna"
+        },
+        {
+          "name": "Lana Del Rey"
+        },
+        {
+          "name": "Katy Perry"
+        },
+        {
+          "name": "Lady Gaga"
+        },
+        {
+          "name": "Miley Cyrus"
+        },
+        {
+          "name": "Mark Ronson"
+        },
+        {
+          "name": "Madonna"
+        },
+        {
+          "name": "Lorde"
+        },
+        {
+          "name": "Khalid"
+        },
+        {
+          "name": "Billie Eilish"
+        },
+        {
+          "name": "Ariana Grande"
+        },
+        {
+          "name": "Taylor Swift"
+        },
+        {
+          "name": "Beyoncé"
+        },
+        {
+          "name": "Shawn Mendes"
+        },
+        {
+          "name": "Rihanna"
+        },
+        {
+          "name": "Lana Del Rey"
+        },
+        {
+          "name": "Katy Perry"
+        },
+        {
+          "name": "Lady Gaga"
+        },
+        {
+          "name": "Miley Cyrus"
+        },
+        {
+          "name": "Mark Ronson"
+        },
+        {
+          "name": "Madonna"
+        },
+        {
+          "name": "Lorde"
+        },
+        {
+          "name": "Khalid"
+        },
+        {
+          "name": "Sia"
+        },
+        {
+          "name": "Sam Smith"
+        },
+        {
+          "name": "Halsey"
+        },
+        {
+          "name": "Michael Jackson"
+        },
+        {
+          "name": "Charli XCX"
+        },
+        {
+          "name": "Britney Spears"
+        },
+        {
+          "name": "Dua Lipa"
+        },
+        {
+          "name": "Jonas Brothers"
+        },
+        {
+          "name": "Bruno Mars"
+        },
+        {
+          "name": "Carly Rae Jepsen"
+        },
+        {
+          "name": "P!nk"
+        },
+        {
+          "name": "Adele"
+        }
+      ],
+      "name": "pop, female vocalists"
+    }
+  ],
+  "name": "Musicians"
+}

algorithms-miscellaneous-3/src/main/resources/kmeans/radial.html

@@ -0,0 +1,68 @@
+<!DOCTYPE html>
+<meta charset="utf-8">
+<style>
+    .node circle {
+        fill: #fff;
+        stroke: steelblue;
+        stroke-width: 1.5px;
+    }
+
+    .node {
+        font: 10px sans-serif;
+    }
+
+    .link {
+        fill: none;
+        stroke: #ccc;
+        stroke-width: 1.5px;
+    }
+</style>
+<body>
+<script src="http://d3js.org/d3.v3.min.js"></script>
+<script>
+    var diameter = 1100;
+    var tree = d3.layout.tree()
+        .size([360, diameter / 2 - 300])
+        .separation(function (a, b) {
+            return (a.parent == b.parent ? 1 : 2) / a.depth;
+        });
+    var diagonal = d3.svg.diagonal.radial()
+        .projection(function (d) {
+            return [d.y, d.x / 180 * Math.PI];
+        });
+    var svg = d3.select("body").append("svg")
+        .attr("width", diameter)
+        .attr("height", diameter - 150)
+        .append("g")
+        .attr("transform", "translate(" + diameter / 2 + "," + diameter / 2 + ")");
+    d3.json("lastfm.json", function (error, root) {
+        var nodes = tree.nodes(root),
+            links = tree.links(nodes);
+        var link = svg.selectAll(".link")
+            .data(links)
+            .enter().append("path")
+            .attr("class", "link")
+            .attr("d", diagonal);
+        var node = svg.selectAll(".node")
+            .data(nodes)
+            .enter().append("g")
+            .attr("class", "node")
+            .attr("transform", function (d) {
+                return "rotate(" + (d.x - 90) + ")translate(" + d.y + ")";
+            })
+        node.append("circle")
+            .attr("r", 4.5);
+        node.append("text")
+            .attr("dy", ".31em")
+            .attr("text-anchor", function (d) {
+                return d.x < 180 ? "start" : "end";
+            })
+            .attr("transform", function (d) {
+                return d.x < 180 ? "translate(8)" : "rotate(180)translate(-8)";
+            })
+            .text(function (d) {
+                return d.name;
+            });
+    });
+    d3.select(self.frameElement).style("height", diameter - 150 + "px");
+</script>