ios - 在 swift 3 中过滤代词、动词、地名

Question

例如:

算法目前可以​​过滤这句话

What is the weather in San Francisco?

对此

什么:代词
是:动词
的:确定者
天气:名词
在:介词
旧金山:地名


我编写的代码使用 Swift 3 并使用 NSLinguisticTagger API。不幸的是，没有太多关于这个特定 API 的文档。

let question = "What is the weather in San Francisco?"
let options: NSLinguisticTagger.Options = [.omitWhitespace, .omitPunctuation, .joinNames]
let schemes = NSLinguisticTagger.availableTagSchemes(forLanguage: "en")
let tagger = NSLinguisticTagger(tagSchemes: schemes, options: Int(options.rawValue))
tagger.string = question
tagger.enumerateTags(in: NSMakeRange(0, (question as NSString).length), scheme: NSLinguisticTagSchemeNameTypeOrLexicalClass, options: options) { (tag, tokenRange, _, _) in
    let token = (question as NSString).substring(with: tokenRange)
    print("\(token): \(tag)")
}



var range = NSRange()
var oldString: String = "Where in the world is Carmen San Diego?"
var newString = (oldString as NSString).replacingCharacters(in: range, with: oldString) as? NSString


var tokenRanges: NSArray?
let tags =  newString?.linguisticTags(
    in: NSMakeRange(0, (question as NSString).length),
    scheme: NSLinguisticTagSchemeNameTypeOrLexicalClass,
    options: options, orthography: nil, tokenRanges: &tokenRanges
)

现在我想过滤名词、动词和地名，以便稍后在此示例天气 API 请求中执行。

is, weather, San Francisco

Answer 1

这似乎可以满足您的要求:

import Foundation

struct TaggedToken {
    let token: String
    let tag: String
}

let question = "What is the weather in San Francisco?"
let options: NSLinguisticTagger.Options = [.omitWhitespace, .omitPunctuation, .joinNames]
let schemes = NSLinguisticTagger.availableTagSchemes(forLanguage: "en")
let tagger = NSLinguisticTagger(tagSchemes: schemes, options: Int(options.rawValue))
tagger.string = question

var nounsVerbsAndPlaceNames = [TaggedToken]()
tagger.enumerateTags(in: NSMakeRange(0, (question as NSString).length), scheme: NSLinguisticTagSchemeNameTypeOrLexicalClass, options: options) { (tag, tokenRange, _, _) in
    let token = (question as NSString).substring(with: tokenRange)
    print("\(token): \(tag)")

    switch tag {
    case "Noun",
         "Verb",
         "PlaceName": nounsVerbsAndPlaceNames.append(TaggedToken(token: token, tag: tag))
    default: break;
    }
}

print(nounsVerbsAndPlaceNames)