我试图在以下步骤中找出一个句子是肯定的还是否定的：
1.)使用斯坦福 NLP 解析器从句子中检索词性（动词、名词、形容词等）。
2.) 使用SentiWordNet找到与每个词性相关的正值和负值。
3.) 将得到的正值和负值相加来计算与句子相关的净正值和净负值。
但问题是，SentiWordNet 根据不同的感觉/上下文返回一个正/负值列表。是否可以将特定句子与词性一起传递给 SentiWordNet 解析器，以便它可以自动判断语义/上下文并仅返回一对正负值？
或者这个问题还有其他替代解决方案吗？
谢谢。
SentoWordNet 演示代码 这可能会对您有所帮助。
// Copyright 2013 Petter Törnberg
//
// This demo code has been kindly provided by Petter Törnberg <pettert@chalmers.se>
// for the SentiWordNet website.
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program 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 General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.util.HashMap;
import java.util.Map;
public class SentiWordNetDemoCode {
private Map<String, Double> dictionary;
public SentiWordNetDemoCode(String pathToSWN) throws IOException {
// This is our main dictionary representation
dictionary = new HashMap<String, Double>();
// From String to list of doubles.
HashMap<String, HashMap<Integer, Double>> tempDictionary = new HashMap<String, HashMap<Integer, Double>>();
BufferedReader csv = null;
try {
csv = new BufferedReader(new FileReader(pathToSWN));
int lineNumber = 0;
String line;
while ((line = csv.readLine()) != null) {
lineNumber++;
// If it's a comment, skip this line.
if (!line.trim().startsWith("#")) {
// We use tab separation
String[] data = line.split("\t");
String wordTypeMarker = data[0];
// Example line:
// POS ID PosS NegS SynsetTerm#sensenumber Desc
// a 00009618 0.5 0.25 spartan#4 austere#3 ascetical#2
// ascetic#2 practicing great self-denial;...etc
// Is it a valid line? Otherwise, through exception.
if (data.length != 6) {
throw new IllegalArgumentException(
"Incorrect tabulation format in file, line: "
+ lineNumber);
}
// Calculate synset score as score = PosS - NegS
Double synsetScore = Double.parseDouble(data[2])
- Double.parseDouble(data[3]);
// Get all Synset terms
String[] synTermsSplit = data[4].split(" ");
// Go through all terms of current synset.
for (String synTermSplit : synTermsSplit) {
// Get synterm and synterm rank
String[] synTermAndRank = synTermSplit.split("#");
String synTerm = synTermAndRank[0] + "#"
+ wordTypeMarker;
int synTermRank = Integer.parseInt(synTermAndRank[1]);
// What we get here is a map of the type:
// term -> {score of synset#1, score of synset#2...}
// Add map to term if it doesn't have one
if (!tempDictionary.containsKey(synTerm)) {
tempDictionary.put(synTerm,
new HashMap<Integer, Double>());
}
// Add synset link to synterm
tempDictionary.get(synTerm).put(synTermRank,
synsetScore);
}
}
}
// Go through all the terms.
for (Map.Entry<String, HashMap<Integer, Double>> entry : tempDictionary
.entrySet()) {
String word = entry.getKey();
Map<Integer, Double> synSetScoreMap = entry.getValue();
// Calculate weighted average. Weigh the synsets according to
// their rank.
// Score= 1/2*first + 1/3*second + 1/4*third ..... etc.
// Sum = 1/1 + 1/2 + 1/3 ...
double score = 0.0;
double sum = 0.0;
for (Map.Entry<Integer, Double> setScore : synSetScoreMap
.entrySet()) {
score += setScore.getValue() / (double) setScore.getKey();
sum += 1.0 / (double) setScore.getKey();
}
score /= sum;
dictionary.put(word, score);
}
} catch (Exception e) {
e.printStackTrace();
} finally {
if (csv != null) {
csv.close();
}
}
}
public double extract(String word, String pos) {
return dictionary.get(word + "#" + pos);
}
public static void main(String [] args) throws IOException {
if(args.length<1) {
System.err.println("Usage: java SentiWordNetDemoCode <pathToSentiWordNetFile>");
return;
}
String pathToSWN = args[0];
SentiWordNetDemoCode sentiwordnet = new SentiWordNetDemoCode(pathToSWN);
System.out.println("good#a "+sentiwordnet.extract("good", "a"));
System.out.println("bad#a "+sentiwordnet.extract("bad", "a"));
System.out.println("blue#a "+sentiwordnet.extract("blue", "a"));
System.out.println("blue#n "+sentiwordnet.extract("blue", "n"));
}
}
我们可以将 pos 传递给 sentiwordnet 解析器。下载模式python模块
from pattern.en import wordnet
print wordnet.synsets("kill",pos="VB")[0].weight
wordnet.synsets 返回同义词列表，我们从中选择第一个项目输出将是（极性，主观性）的元组希望这有帮助......