RegressionModel.java

package edu.northwestern.cbits.purple_robot_manager.models;

import java.util.ArrayList;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Map;
import java.util.Set;
import java.util.regex.Pattern;

import android.content.Context;
import android.net.Uri;
import edu.northwestern.cbits.purple_robot_manager.R;
import edu.northwestern.cbits.purple_robot_manager.util.Slugify;

public class RegressionModel extends TrainedModel
{
    public static final String TYPE = "linear-regression";

    private double _intercept = 0.0;

    private Map<String, Double> _coefficients = new HashMap<>();
    private Set<String> _variables = new HashSet<>();

    public RegressionModel(Context context, Uri uri)
    {
        super(context, uri);
    }

    protected void generateModel(Context context, Object modelString)
    {
        /*
         * 08-01 11:39:30.163: E/PR(15876): undefined_feature_value_dt_calmr1 =
         * 08-01 11:39:30.163: E/PR(15876): 0.4026 *
         * networkprobe_hostname=?,android-a6e300e929cef636 + 08-01
         * 11:39:30.163: E/PR(15876): 0.4026 *
         * networkprobe_hostname=android-a6e300e929cef636 + 08-01 11:39:30.163:
         * E/PR(15876): 0.4026 * networkprobe_ip_address=?,192168121 + 08-01
         * 11:39:30.163: E/PR(15876): 0.4026 * networkprobe_ip_address=192168121
         * + 08-01 11:39:30.163: E/PR(15876): -2.8561 *
         * robothealthprobe_cpu_usage + 08-01 11:39:30.163: E/PR(15876): -1.1707
         * * telephonyprobe_call_state + 08-01 11:39:30.163: E/PR(15876): 0.286
         * * wifiaccesspointsprobe_access_point_count + 08-01 11:39:30.163:
         * E/PR(15876): 0.4026 *
         * wifiaccesspointsprobe_current_bssid=?,640f28c1da89 + 08-01
         * 11:39:30.163: E/PR(15876): 0.0191 *
         * wifiaccesspointsprobe_current_rssi + 08-01 11:39:30.163: E/PR(15876):
         * 0.4026 * wifiaccesspointsprobe_current_ssid=allison-and-alex + 08-01
         * 11:39:30.163: E/PR(15876): 2.3794
         */

        for (String line : modelString.toString().split("[\\r\\n]+"))
        {
            if (line.startsWith("  "))
            {
                try
                {
                    this._intercept = Double.parseDouble(line);
                }
                catch (NumberFormatException e)
                {
                    String[] toks = line.split("\\*");

                    String coef = toks[0].trim();
                    String args = toks[1].trim();

                    int index = line.indexOf("=");

                    if (index == -1)
                    {
                        args = args.replaceAll(Pattern.quote("+"), "").trim();
                        this._coefficients.put(args, Double.parseDouble(coef));

                        this._variables.add(args);
                    }
                    else
                    {
                        String[] argToks = args.split("=");

                        String probe = argToks[0].trim();
                        this._variables.add(probe);

                        ArrayList<String> values = new ArrayList<>();

                        String probeValues = argToks[1].trim();

                        if (probeValues.contains(",") == false)
                            values.add(probeValues);
                        else
                        {
                            String[] valueToks = probeValues.split(",");

                            for (String tok : valueToks)
                            {
                                values.add(tok.trim());
                            }
                        }

                        for (String value : values)
                        {
                            value = value.replaceAll(Pattern.quote("+"), "").trim();

                            String coefKey = probe + "_" + value;

                            this._coefficients.put(coefKey, Double.parseDouble(coef));
                        }
                    }
                }
            }
        }
    }

    protected Object evaluateModel(Context context, Map<String, Object> snapshot)
    {
        ArrayList<String> requiredKeys = new ArrayList<>();
        requiredKeys.addAll(this._variables);

        double prediction = this._intercept;

        for (String key : snapshot.keySet())
        {
            Object value = snapshot.get(key);

            if (requiredKeys.contains(key))
            {
                Double coef = this._coefficients.get(key);

                double valueValue = 1.0;

                if (value instanceof Double || value instanceof Float || value instanceof Long
                        || value instanceof Integer)
                    valueValue = Double.parseDouble("" + value);

                if (coef == null)
                {
                    coef = this._coefficients.get(key + "_" + Slugify.slugify(value.toString()));

                    if (coef == null)
                    {
                        coef = this._coefficients.get(key + "_?");

                        if (coef == null)
                            coef = 0.0;
                    }
                }

                prediction += (coef * valueValue);

                requiredKeys.remove(key);
            }
        }

        return prediction;
    }

    public String modelType()
    {
        return RegressionModel.TYPE;
    }

    public String summary(Context context)
    {
        return context.getString(R.string.summary_model_regression);
    }
}