Click here to Skip to main content
15,883,857 members
Articles / Containers / Virtual Machine

Statistical parsing of English sentences

Rate me:
Please Sign up or sign in to vote.
4.95/5 (88 votes)
13 Dec 200617 min read 986.2K   23.6K   239  
Shows how to generate parse trees for English language sentences, using a C# port of OpenNLP, a statistical natural language parsing library.
//Copyright (C) 2005 Richard J. Northedge
//
// This library is free software; you can redistribute it and/or
// modify it under the terms of the GNU Lesser General Public
// License as published by the Free Software Foundation; either
// version 2.1 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 Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public
// License along with this program; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.

//This file is based on the BuildContextGenerator.java source file found in the
//original java implementation of OpenNLP.  That source file contains the following header:

//Copyright (C) 2003 Thomas Morton
// 
//This library is free software; you can redistribute it and/or
//modify it under the terms of the GNU Lesser General Public
//License as published by the Free Software Foundation; either
//version 2.1 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 Lesser General Public License for more details.
// 
//You should have received a copy of the GNU Lesser General Public
//License along with this program; if not, write to the Free Software
//Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.

using System;
using System.Collections.Generic;
using System.Text;

namespace OpenNLP.Tools.Parser
{
	/// <summary>
	/// Class to generator predictive contexts for deciding how constituents should be combined together.
	/// </summary>
	public class BuildContextGenerator : SharpEntropy.IContextGenerator
	{
		
		private const string mEndOfSentence = "eos";
		
		/// <summary> 
		/// Creates a new context generator for making decisions about combining constituents together.
		/// </summary>
		public BuildContextGenerator()
		{
		}

        /// <summary>
        /// Returns the predictive context used to determine how the constituent at the specified index 
        /// should be combined with other constituents. 
        /// </summary>
        /// <param name="input">
        /// Object array containing an array of Parse objects (the uncombined constituents), and the index of the
        /// relevant constituent.
        /// </param>
		public virtual string[] GetContext(object input)
		{
			object[] parameters = (object[]) input;
			return GetContext((Parse[]) parameters[0], ((int) parameters[1]));
		}
		
		private string MakeConstituent(Parse inputParse, int index)
		{
			StringBuilder feature = new StringBuilder(20);
			feature.Append(index).Append("=");
			if (inputParse != null)
			{
				if (index < 0)
				{
					feature.Append(inputParse.Label).Append("|");
				}
				feature.Append(inputParse.Type).Append("|").Append(inputParse.Head.ToString());
			}
			else
			{
				feature.Append(mEndOfSentence).Append("|").Append(mEndOfSentence).Append("|").Append(mEndOfSentence);
			}
			return feature.ToString();
		}
		
		private string MakeConstituentBackOff(Parse inputParse, int index)
		{
			StringBuilder feature = new StringBuilder(20);
			feature.Append(index).Append("*=");
			if (inputParse != null)
			{
				if (index < 0)
				{
					feature.Append(inputParse.Label).Append("|");
				}
				feature.Append(inputParse.Type);
			}
			else
			{
				feature.Append(mEndOfSentence).Append("|").Append(mEndOfSentence);
			}
			return feature.ToString();
		}
		
		/// <summary>
		/// Returns the predictive context used to determine how the constituent at the specified index 
        /// should be combined with other constituents. 
		/// </summary>
		/// <param name="constituents">
		/// The constituents which have yet to be combined into new constituents.
		/// </param>
		/// <param name="index">
		/// The index of the constituent whcihi is being considered.
		/// </param>
		/// <returns>
		/// the context for building constituents at the specified index.
		/// </returns>
		public virtual string[] GetContext(Parse[] constituents, int index)
		{
			List<string> features = new List<string>(100);
			int constituentCount = constituents.Length;
			
			//default 
			features.Add("default");
			// cons(-2), cons(-1), cons(0), cons(1), cons(2)
			// cons(-2)
			Parse previousPreviousParse = null;
			Parse previousParse = null;
			Parse currentParse = null;
			Parse nextParse = null;
			Parse nextNextParse = null;
			
			if (index - 2 >= 0)
			{
				previousPreviousParse = constituents[index - 2];
			}
			if (index - 1 >= 0)
			{
				previousParse = constituents[index - 1];
			}
			currentParse = constituents[index];
			if (index + 1 < constituentCount)
			{
				nextParse = constituents[index + 1];
			}
			if (index + 2 < constituentCount)
			{
				nextNextParse = constituents[index + 2];
			}
			
			// cons(-2), cons(-1), cons(0), cons(1), cons(2)
			string previousPreviousConstituent = MakeConstituent(previousPreviousParse, - 2);
			string previousConstituent = MakeConstituent(previousParse, - 1);
			string currentConstituent = MakeConstituent(currentParse, 0);
			string nextConstituent = MakeConstituent(nextParse, 1);
			string nextNextConstituent = MakeConstituent(nextNextParse, 2);
			
			string previousPreviousConstituentBackOff = MakeConstituentBackOff(previousPreviousParse, - 2);
			string previousConstituentBackOff = MakeConstituentBackOff(previousParse, - 1);
			string currentConstituentBackOff = MakeConstituentBackOff(currentParse, 0);
			string nextConstituentBackOff = MakeConstituentBackOff(nextParse, 1);
			string nextNextConstituentBackOff = MakeConstituentBackOff(nextNextParse, 2);
			
			// cons(-2), cons(-1), cons(0), cons(1), cons(2)
			features.Add(previousPreviousConstituent);
			features.Add(previousPreviousConstituentBackOff);
			features.Add(previousConstituent);
			features.Add(previousConstituentBackOff);
			features.Add(currentConstituent);
			features.Add(currentConstituentBackOff);
			features.Add(nextConstituent);
			features.Add(nextConstituentBackOff);
			features.Add(nextNextConstituent);
			features.Add(nextNextConstituentBackOff);
			
			// cons(-1,0), cons(0,1)
			features.Add(previousConstituent + "," + currentConstituent);
			features.Add(previousConstituentBackOff + "," + currentConstituent);
			features.Add(previousConstituent + "," + currentConstituentBackOff);
			features.Add(previousConstituentBackOff + "," + currentConstituentBackOff);
			
			features.Add(currentConstituent + "," + nextConstituent);
			features.Add(currentConstituentBackOff + "," + nextConstituent);
			features.Add(currentConstituent + "," + nextConstituentBackOff);
			features.Add(currentConstituentBackOff + "," + nextConstituentBackOff);
			
			// cons3(-2,-1,0), cons3(-1,0,1), cons3(0,1,2)
			features.Add(previousPreviousConstituent + "," + previousConstituent + "," + currentConstituent);
			features.Add(previousPreviousConstituentBackOff + "," + previousConstituent + "," + currentConstituent);
			features.Add(previousPreviousConstituent + "," + previousConstituentBackOff + "," + currentConstituent);
			features.Add(previousPreviousConstituentBackOff + "," + previousConstituentBackOff + "," + currentConstituent);
			features.Add(previousPreviousConstituentBackOff + "," + previousConstituentBackOff + "," + currentConstituentBackOff);
			
			features.Add(previousConstituent + "," + currentConstituent + "," + nextConstituent);
			features.Add(previousConstituentBackOff + "," + currentConstituent + "," + nextConstituent);
			features.Add(previousConstituent + "," + currentConstituent + "," + nextConstituentBackOff);
			features.Add(previousConstituentBackOff + "," + currentConstituent + "," + nextConstituentBackOff);
			features.Add(previousConstituentBackOff + "," + currentConstituentBackOff + "," + nextConstituentBackOff);
			
			features.Add(currentConstituent + "," + nextConstituent + "," + nextNextConstituent);
			features.Add(currentConstituent + "," + nextConstituentBackOff + "," + nextNextConstituent);
			features.Add(currentConstituent + "," + nextConstituent + "," + nextNextConstituentBackOff);
			features.Add(currentConstituent + "," + nextConstituentBackOff + "," + nextNextConstituentBackOff);
			features.Add(currentConstituentBackOff + "," + nextConstituentBackOff + "," + nextNextConstituentBackOff);
			
			// punct
			string currentParseWord = currentParse.ToString();
			if (currentParseWord == "-RRB-")
			{
				for (int parseIndex = index - 1; parseIndex >= 0; parseIndex--)
				{
					Parse testParse = constituents[parseIndex];
					if (testParse.ToString() == "-LRB-")
					{
						features.Add("bracketsmatch");
						break;
					}
					if (testParse.Label.StartsWith(MaximumEntropyParser.StartPrefix))
					{
						break;
					}
				}
			}
			if (currentParseWord == "-RCB-")
			{
				for (int parseIndex = index - 1; parseIndex >= 0; parseIndex--)
				{
					Parse testParse = constituents[parseIndex];
					if (testParse.ToString() == "-LCB-")
					{
						features.Add("bracketsmatch");
						break;
					}
					if (testParse.Label.StartsWith(MaximumEntropyParser.StartPrefix))
					{
						break;
					}
				}
			}
			if (currentParseWord == "''")
			{
				for (int parseIndex = index - 1; parseIndex >= 0; parseIndex--)
				{
					Parse testParse = constituents[parseIndex];
					if (testParse.ToString() == "``")
					{
						features.Add("quotesmatch");
						break;
					}
					if (testParse.Label.StartsWith(MaximumEntropyParser.StartPrefix))
					{
						break;
					}
				}
			}
			if (currentParseWord == "'")
			{
				for (int parseIndex = index - 1; parseIndex >= 0; parseIndex--)
				{
					Parse testParse = constituents[parseIndex];
					if (testParse.ToString() == "`")
					{
						features.Add("quotesmatch");
						break;
					}
					if (testParse.Label.StartsWith(MaximumEntropyParser.StartPrefix))
					{
						break;
					}
				}
			}
			if (currentParseWord == ",")
			{
				for (int parseIndex = index - 1; parseIndex >= 0; parseIndex--)
				{
					Parse testParse = constituents[parseIndex];
					if (testParse.ToString() == ",")
					{
						features.Add("iscomma");
						break;
					}
					if (testParse.Label.StartsWith(MaximumEntropyParser.StartPrefix))
					{
						break;
					}
				}
			}
			if (currentParseWord == (".") && index == constituentCount - 1)
			{
				for (int parseIndex = index - 1; parseIndex >= 0; parseIndex--)
				{
					Parse testParse = constituents[parseIndex];
					if (testParse.Label.StartsWith(MaximumEntropyParser.StartPrefix))
					{
						if (parseIndex == 0)
						{
							features.Add("endofsentence");
						}
						break;
					}
				}
			}
			return features.ToArray();
		}
	}
}

By viewing downloads associated with this article you agree to the Terms of Service and the article's licence.

If a file you wish to view isn't highlighted, and is a text file (not binary), please let us know and we'll add colourisation support for it.

License

This article has no explicit license attached to it but may contain usage terms in the article text or the download files themselves. If in doubt please contact the author via the discussion board below.

A list of licenses authors might use can be found here


Written By
Web Developer
United Kingdom United Kingdom
Richard Northedge is a senior developer with a UK Microsoft Gold Partner company. He has a postgraduate degree in English Literature, has been programming professionally since 1998 and has been an MCSD since 2000.

Comments and Discussions