/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.lucene.search.similarities;


import java.util.ArrayList;
import java.util.List;
import java.util.Locale;

import org.apache.lucene.search.Explanation;

Language model based on the Jelinek-Mercer smoothing method. From Chengxiang Zhai and John Lafferty. 2001. A study of smoothing methods for language models applied to Ad Hoc information retrieval. In Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval (SIGIR '01). ACM, New York, NY, USA, 334-342.

The model has a single parameter, λ. According to said paper, the optimal value depends on both the collection and the query. The optimal value is around 0.1 for title queries and 0.7 for long queries.

Values should be between 0 (exclusive) and 1 (inclusive). Values near zero act score more like a conjunction (coordinate level matching), whereas values near 1 behave the opposite (more like pure disjunction).

@lucene.experimental
/** * Language model based on the Jelinek-Mercer smoothing method. From Chengxiang * Zhai and John Lafferty. 2001. A study of smoothing methods for language * models applied to Ad Hoc information retrieval. In Proceedings of the 24th * annual international ACM SIGIR conference on Research and development in * information retrieval (SIGIR '01). ACM, New York, NY, USA, 334-342. * <p>The model has a single parameter, &lambda;. According to said paper, the * optimal value depends on both the collection and the query. The optimal value * is around {@code 0.1} for title queries and {@code 0.7} for long queries.</p> * <p>Values should be between 0 (exclusive) and 1 (inclusive). Values near zero act score more * like a conjunction (coordinate level matching), whereas values near 1 behave * the opposite (more like pure disjunction). * @lucene.experimental */
public class LMJelinekMercerSimilarity extends LMSimilarity {
The λ parameter.
/** The &lambda; parameter. */
private final float lambda;
Instantiates with the specified collectionModel and λ parameter.
/** Instantiates with the specified collectionModel and &lambda; parameter. */
public LMJelinekMercerSimilarity( CollectionModel collectionModel, float lambda) { super(collectionModel); if (Float.isNaN(lambda) || lambda <= 0 || lambda > 1) { throw new IllegalArgumentException("lambda must be in the range (0 .. 1]"); } this.lambda = lambda; }
Instantiates with the specified λ parameter.
/** Instantiates with the specified &lambda; parameter. */
public LMJelinekMercerSimilarity(float lambda) { if (Float.isNaN(lambda) || lambda <= 0 || lambda > 1) { throw new IllegalArgumentException("lambda must be in the range (0 .. 1]"); } this.lambda = lambda; } @Override protected double score(BasicStats stats, double freq, double docLen) { return stats.getBoost() * Math.log(1 + ((1 - lambda) * freq / docLen) / (lambda * ((LMStats)stats).getCollectionProbability())); } @Override protected void explain(List<Explanation> subs, BasicStats stats, double freq, double docLen) { if (stats.getBoost() != 1.0d) { subs.add(Explanation.match((float) stats.getBoost(), "boost")); } subs.add(Explanation.match(lambda, "lambda")); double p = ((LMStats)stats).getCollectionProbability(); Explanation explP = Explanation.match((float) p, "P, probability that the current term is generated by the collection"); subs.add(explP); Explanation explFreq = Explanation.match((float) freq, "freq, number of occurrences of term in the document"); subs.add(explFreq); subs.add(Explanation.match((float) docLen,"dl, length of field")); super.explain(subs, stats, freq, docLen); } @Override protected Explanation explain( BasicStats stats, Explanation freq, double docLen) { List<Explanation> subs = new ArrayList<>(); explain(subs, stats, freq.getValue().doubleValue(), docLen); return Explanation.match( (float) score(stats, freq.getValue().doubleValue(), docLen), "score(" + getClass().getSimpleName() + ", freq=" + freq.getValue() +"), computed as boost * " + "log(1 + ((1 - lambda) * freq / dl) /(lambda * P)) from:", subs); }
Returns the λ parameter.
/** Returns the &lambda; parameter. */
public float getLambda() { return lambda; } @Override public String getName() { return String.format(Locale.ROOT, "Jelinek-Mercer(%f)", getLambda()); } }