409 lines
13 KiB
PHP
409 lines
13 KiB
PHP
<?php
|
||
/*
|
||
* Copyright (c) 2017 Baidu.com, Inc. All Rights Reserved
|
||
*
|
||
* Licensed under the Apache License, Version 2.0 (the "License"); you may not
|
||
* use this file except in compliance with the License. You may obtain a copy of
|
||
* the License at
|
||
*
|
||
* Http://www.apache.org/licenses/LICENSE-2.0
|
||
*
|
||
* Unless required by applicable law or agreed to in writing, software
|
||
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
||
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
||
* License for the specific language governing permissions and limitations under
|
||
* the License.
|
||
*/
|
||
|
||
require_once 'lib/AipBase.php';
|
||
class AipNlp extends AipBase {
|
||
|
||
/**
|
||
* 词法分析 lexer api url
|
||
* @var string
|
||
*/
|
||
private $lexerUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/lexer';
|
||
|
||
/**
|
||
* 词法分析（定制版） lexer_custom api url
|
||
* @var string
|
||
*/
|
||
private $lexerCustomUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/lexer_custom';
|
||
|
||
/**
|
||
* 依存句法分析 dep_parser api url
|
||
* @var string
|
||
*/
|
||
private $depParserUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/depparser';
|
||
|
||
/**
|
||
* 词向量表示 word_embedding api url
|
||
* @var string
|
||
*/
|
||
private $wordEmbeddingUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v2/word_emb_vec';
|
||
|
||
/**
|
||
* DNN语言模型 dnnlm_cn api url
|
||
* @var string
|
||
*/
|
||
private $dnnlmCnUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v2/dnnlm_cn';
|
||
|
||
/**
|
||
* 词义相似度 word_sim_embedding api url
|
||
* @var string
|
||
*/
|
||
private $wordSimEmbeddingUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v2/word_emb_sim';
|
||
|
||
/**
|
||
* 短文本相似度 simnet api url
|
||
* @var string
|
||
*/
|
||
private $simnetUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v2/simnet';
|
||
|
||
/**
|
||
* 评论观点抽取 comment_tag api url
|
||
* @var string
|
||
*/
|
||
private $commentTagUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v2/comment_tag';
|
||
|
||
/**
|
||
* 情感倾向分析 sentiment_classify api url
|
||
* @var string
|
||
*/
|
||
private $sentimentClassifyUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/sentiment_classify';
|
||
|
||
/**
|
||
* 文章标签 keyword api url
|
||
* @var string
|
||
*/
|
||
private $keywordUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/keyword';
|
||
|
||
/**
|
||
* 文章分类 topic api url
|
||
* @var string
|
||
*/
|
||
private $topicUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/topic';
|
||
|
||
/**
|
||
* 文本纠错 ecnet api url
|
||
* @var string
|
||
*/
|
||
private $ecnetUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/ecnet';
|
||
|
||
/**
|
||
* 对话情绪识别接口 emotion api url
|
||
* @var string
|
||
*/
|
||
private $emotionUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/emotion';
|
||
|
||
/**
|
||
* 新闻摘要接口 news_summary api url
|
||
* @var string
|
||
*/
|
||
private $newsSummaryUrl = 'https://aip.baidubce.com/rpc/2.0/nlp/v1/news_summary';
|
||
|
||
/**
|
||
* 格式化结果
|
||
* @param $content string
|
||
* @return mixed
|
||
*/
|
||
protected function proccessResult($content){
|
||
return json_decode(mb_convert_encoding($content, 'UTF8', 'GBK'), true, 512, JSON_BIGINT_AS_STRING);
|
||
}
|
||
|
||
/**
|
||
* 词法分析接口
|
||
*
|
||
* @param string $text - 待分析文本（目前仅支持GBK编码），长度不超过65536字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function lexer($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->lexerUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 词法分析（定制版）接口
|
||
*
|
||
* @param string $text - 待分析文本（目前仅支持GBK编码），长度不超过65536字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function lexerCustom($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->lexerCustomUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 依存句法分析接口
|
||
*
|
||
* @param string $text - 待分析文本（目前仅支持GBK编码），长度不超过256字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* mode 模型选择。默认值为0，可选值mode=0（对应web模型）；mode=1（对应query模型）
|
||
* @return array
|
||
*/
|
||
public function depParser($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->depParserUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 词向量表示接口
|
||
*
|
||
* @param string $word - 文本内容（GBK编码），最大64字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function wordEmbedding($word, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['word'] = $word;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->wordEmbeddingUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* DNN语言模型接口
|
||
*
|
||
* @param string $text - 文本内容（GBK编码），最大512字节，不需要切词
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function dnnlm($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->dnnlmCnUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 词义相似度接口
|
||
*
|
||
* @param string $word1 - 词1（GBK编码），最大64字节
|
||
* @param string $word2 - 词1（GBK编码），最大64字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* mode 预留字段，可选择不同的词义相似度模型。默认值为0，目前仅支持mode=0
|
||
* @return array
|
||
*/
|
||
public function wordSimEmbedding($word1, $word2, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['word_1'] = $word1;
|
||
$data['word_2'] = $word2;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->wordSimEmbeddingUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 短文本相似度接口
|
||
*
|
||
* @param string $text1 - 待比较文本1（GBK编码），最大512字节
|
||
* @param string $text2 - 待比较文本2（GBK编码），最大512字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* model 默认为"BOW"，可选"BOW"、"CNN"与"GRNN"
|
||
* @return array
|
||
*/
|
||
public function simnet($text1, $text2, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text_1'] = $text1;
|
||
$data['text_2'] = $text2;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->simnetUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 评论观点抽取接口
|
||
*
|
||
* @param string $text - 评论内容（GBK编码），最大10240字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* type 评论行业类型，默认为4（餐饮美食）
|
||
* @return array
|
||
*/
|
||
public function commentTag($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->commentTagUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 情感倾向分析接口
|
||
*
|
||
* @param string $text - 文本内容（GBK编码），最大102400字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function sentimentClassify($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->sentimentClassifyUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 文章标签接口
|
||
*
|
||
* @param string $title - 篇章的标题，最大80字节
|
||
* @param string $content - 篇章的正文，最大65535字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function keyword($title, $content, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['title'] = $title;
|
||
$data['content'] = $content;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->keywordUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 文章分类接口
|
||
*
|
||
* @param string $title - 篇章的标题，最大80字节
|
||
* @param string $content - 篇章的正文，最大65535字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function topic($title, $content, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['title'] = $title;
|
||
$data['content'] = $content;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->topicUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 文本纠错接口
|
||
*
|
||
* @param string $text - 待纠错文本，输入限制511字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* @return array
|
||
*/
|
||
public function ecnet($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->ecnetUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 对话情绪识别接口接口
|
||
*
|
||
* @param string $text - 待识别情感文本，输入限制512字节
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* scene default（默认项-不区分场景），talk（闲聊对话-如度秘聊天等），task（任务型对话-如导航对话等），customer_service（客服对话-如电信/银行客服等）
|
||
* @return array
|
||
*/
|
||
public function emotion($text, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['text'] = $text;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->emotionUrl, $data);
|
||
}
|
||
|
||
/**
|
||
* 新闻摘要接口接口
|
||
*
|
||
* @param string $content - 字符串（限3000字符数以内）字符串仅支持GBK编码，长度需小于3000字符数（即6000字节），请输入前确认字符数没有超限，若字符数超长会返回错误。正文中如果包含段落信息，请使用"\n"分隔，段落信息算法中有重要的作用，请尽量保留
|
||
* @param integer $maxSummaryLen - 此数值将作为摘要结果的最大长度。例如：原文长度1000字，本参数设置为150，则摘要结果的最大长度是150字；推荐最优区间：200-500字
|
||
* @param array $options - 可选参数对象，key: value都为string类型
|
||
* @description options列表:
|
||
* title 字符串（限200字符数）字符串仅支持GBK编码，长度需小于200字符数（即400字节），请输入前确认字符数没有超限，若字符数超长会返回错误。标题在算法中具有重要的作用，若文章确无标题，输入参数的“标题”字段为空即可
|
||
* @return array
|
||
*/
|
||
public function newsSummary($content, $maxSummaryLen, $options=array()){
|
||
|
||
$data = array();
|
||
|
||
$data['content'] = $content;
|
||
$data['max_summary_len'] = $maxSummaryLen;
|
||
|
||
$data = array_merge($data, $options);
|
||
$data = mb_convert_encoding(json_encode($data), 'GBK', 'UTF8');
|
||
|
||
return $this->request($this->newsSummaryUrl, $data);
|
||
}
|
||
} |