summarizer.h

Go to the documentation of this file.

//
//    FreeLing - Open Source Language Analyzers
//
//    Copyright (C) 2014   TALP Research Center
//                         Universitat Politecnica de Catalunya
//
//    This library is free software; you can redistribute it and/or
//    modify it under the terms of the GNU Affero General Public
//    License as published by the Free Software Foundation; either
//    version 3 of the License, or (at your option) any later version.
//
//    This library is distributed in the hope that it will be useful,
//    but WITHOUT ANY WARRANTY; without even the implied warranty of
//    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
//    Affero General Public License for more details.
//
//    You should have received a copy of the GNU Affero General Public
//    License along with this library; if not, write to the Free Software
//    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
//
//    contact: Lluis Padro (padro@lsi.upc.es)
//             TALP Research Center
//             despatx C6.212 - Campus Nord UPC
//             08034 Barcelona.  SPAIN
//

#ifndef _SUMMARIZER
#define _SUMMARIZER

#include "freeling/windll.h"
#include "freeling/morfo/lexical_chain.h"
#include "freeling/morfo/language.h"

namespace freeling {


  class WINDLL summarizer {
  public:
    summarizer(const std::wstring &datFile);

    ~summarizer();

    std::list<const freeling::sentence*> summarize(const freeling::document &doc,
                                                   int num_words) const;

    typedef enum {FIRST_WORD,FIRST_MOST_WEIGHT,WEIGHT_SUM} Heuristics;

  private:

    bool remove_used_lexical_chains;
    bool only_strong;
    int hypernymy_depth;
    double alpha;
    std::wstring semdb_path;
    std::set<relation*> used_relations;
    Heuristics heuristic;

    std::map<std::wstring, std::list<lexical_chain>> build_lexical_chains(const freeling::document &doc) const;

    void remove_one_word_lexical_chains(std::map<std::wstring, std::list<lexical_chain>> &chains) const;

    void remove_weak_lexical_chains(std::map<std::wstring, std::list<lexical_chain>> &chains) const;
    
    void print_lexical_chains(std::map<std::wstring, std::list<lexical_chain>> &chains) const;
    
    int count_occurences(const freeling::word &w, const freeling::document &doc) const;
    
    double average_scores(std::map<std::wstring, std::list<lexical_chain> > &chains_type) const;
    
    double standard_deviation_scores(std::map<std::wstring, std::list<lexical_chain> > &chains_type,
                                     const double avg) const;
    
    std::list<lexical_chain> map_to_lists(std::map<std::wstring,
                                          std::list<lexical_chain> > &chains_type) const;
    
    void compute_sentence(const std::list<word_pos> &wps, std::list<word_pos> &wp_list,
                          std::set<const freeling::sentence*> &sent_set, int &acc_n_words,
                          int num_words) const;
    
    std::list<word_pos> first_word(std::map<std::wstring,
                                   std::list<lexical_chain> > &chains_type, int num_words) const;
    
    std:: list<word_pos> first_most_weighted_word(std::map<std::wstring, 
                                                  std::list<lexical_chain> > &chains, 
                                                  int num_words) const;
    
    std::list<word_pos> sum_of_chain_weights(std::map<std::wstring, 
                                             std::list<lexical_chain> > &chains, 
                                             int num_words) const;

  };

} // namespace

#endif