xapian-letor/feature/tfdoclencolltfcolllenfeature.cc

   1 /** @file tfdoclencolltfcolllenfeature.cc
   2  * @brief TfDoclenCollTfCollLenFeature class
   3  */
   4 /* Copyright (C) 2012 Parth Gupta
   5  * Copyright (C) 2016 Ayush Tomar
   6  *
   7  * This program is free software; you can redistribute it and/or
   8  * modify it under the terms of the GNU General Public License as
   9  * published by the Free Software Foundation; either version 2 of the
  10  * License, or (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  15  * GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License
  18  * along with this program; if not, write to the Free Software
  19  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
  20  */
  21
  22 #include <config.h>
  23
  24 #include "xapian-letor/feature.h"
  25
  26 #include "debuglog.h"
  27 #include "stringutils.h"
  28
  29 using namespace std;
  30
  31 namespace Xapian {
  32
  33 string
  34 TfDoclenCollTfCollLenFeature::name() const
  35 {
  36     return "TfDoclenCollTfCollLenFeature";
  37 }
  38
  39 /** A helper function for feature->get_value()
  40  *
  41  *  Checks if the term belongs to the title or is stemmed from the title.
  42  */
  43 inline bool
  44 is_title_term(const std::string& term)
  45 {
  46     return startswith(term, 'S') || startswith(term, "ZS");
  47 }
  48
  49 vector<double>
  50 TfDoclenCollTfCollLenFeature::get_values() const
  51 {
  52     LOGCALL(API, vector<double>, "TfDoclenCollTfCollLenFeature::get_values", NO_ARGS);
  53
  54     vector<double> values;
  55     double value = 0;
  56     double coll_len;
  57     double doc_len;
  58     auto coll_len_iterator = collection_length.find("title");
  59     if (coll_len_iterator != collection_length.end())
  60         coll_len = (double)coll_len_iterator->second;
  61     else
  62         coll_len = 0;
  63     auto doc_len_iterator = doc_length.find("title");
  64     if (doc_len_iterator != doc_length.end())
  65         doc_len = (double)doc_len_iterator->second;
  66     else
  67         doc_len = 0;
  68
  69     for (TermIterator qt = feature_query.get_unique_terms_begin();
  70          qt != feature_query.get_terms_end(); ++qt) {
  71         if (is_title_term((*qt))) {
  72             double tf;
  73             double coll_tf;
  74             auto tf_iterator = termfreq.find(*qt);
  75             auto coll_tf_iterator = collection_termfreq.find(*qt);
  76             if (tf_iterator != termfreq.end())
  77                 tf = (double)tf_iterator->second;
  78             else
  79                 tf = 0;
  80             if (coll_tf_iterator != collection_termfreq.end())
  81                 coll_tf = (double)coll_tf_iterator->second;
  82             else
  83                 coll_tf = 0;
  84             value += log10(1 + ((tf * coll_len) / (1 + (doc_len * coll_tf))));
  85         }
  86     }
  87     values.push_back(value);
  88     value = 0;
  89     coll_len_iterator = collection_length.find("body");
  90     if (coll_len_iterator != collection_length.end())
  91         coll_len = (double)coll_len_iterator->second;
  92     else
  93         coll_len = 0;
  94     doc_len_iterator = doc_length.find("body");
  95     if (doc_len_iterator != doc_length.end())
  96         doc_len = (double)doc_len_iterator->second;
  97     else
  98         doc_len = 0;
  99
 100     for (Xapian::TermIterator qt = feature_query.get_unique_terms_begin();
 101          qt != feature_query.get_terms_end(); ++qt) {
 102         if (!is_title_term((*qt))) {
 103             double tf;
 104             double coll_tf;
 105             auto tf_iterator = termfreq.find(*qt);
 106             auto coll_tf_iterator = collection_termfreq.find(*qt);
 107             if (tf_iterator != termfreq.end())
 108                 tf = (double)tf_iterator->second;
 109             else
 110                 tf = 0;
 111             if (coll_tf_iterator != collection_termfreq.end())
 112                 coll_tf = (double)coll_tf_iterator->second;
 113             else
 114                 coll_tf = 0;
 115             value += log10(1 + ((tf * coll_len) / (1 + (doc_len * coll_tf))));
 116         }
 117     }
 118     values.push_back(value);
 119     value = 0;
 120     coll_len_iterator = collection_length.find("whole");
 121     if (coll_len_iterator != collection_length.end())
 122         coll_len = (double)coll_len_iterator->second;
 123     else
 124         coll_len = 0;
 125     doc_len_iterator = doc_length.find("whole");
 126     if (doc_len_iterator != doc_length.end())
 127         doc_len = (double)doc_len_iterator->second;
 128     else
 129         doc_len = 0;
 130
 131     for (Xapian::TermIterator qt = feature_query.get_unique_terms_begin();
 132          qt != feature_query.get_terms_end(); ++qt) {
 133         double tf;
 134         double coll_tf;
 135         auto tf_iterator = termfreq.find(*qt);
 136         auto coll_tf_iterator = collection_termfreq.find(*qt);
 137         if (tf_iterator != termfreq.end())
 138             tf = (double)tf_iterator->second;
 139         else
 140             tf = 0;
 141         if (coll_tf_iterator != collection_termfreq.end())
 142             coll_tf = (double)coll_tf_iterator->second;
 143         else
 144             coll_tf = 0;
 145         value += log10(1 + ((tf * coll_len) / (1 + (doc_len * coll_tf))));
 146     }
 147     values.push_back(value);
 148
 149     return values;
 150 }
 151
 152 }