The following is the request body I sent:
"suggest":{
"text":"运动写", //Chinese
"simple_term":{
"term":{
"analyzer":"simple_pinyin_analyzer", //use whitespace tokenizer and pinyin filter
"field":"keywords.pinyin", //use ik_smart tokenizer and pinyin filter
"suggest_mode":"ALWAYS",
"accuracy":0.5,
"sort":"SCORE",
"string_distance":"INTERNAL",
"max_edits":2,
"max_inspections":5,
"max_term_freq":0.01,
"prefix_length":1,
"min_word_length":2,
"min_doc_freq":0
}
}
The results:
"simple_term": [
{
"text": "运动写",
"offset": 0,
"length": 3,
"options": [
{
"text": "运动鞋",
"score": 0.6666666,
"freq": 1
}
]
},
{
"text": "yundongxie",
"offset": 0,
"length": 3,
"options": []
},
{
"text": "ydx",
"offset": 0,
"length": 3,
"options": []
}
]
At the same time split into Chinese and Pinyin, respectively, for error correction, Chinese can get error correction results (some error correction results are not ideal), for Pinyin error correction, due to correct spelling, so no results. The purpose that I want to achieve is to match Chinese according to pinyin and return the corresponding Chinese result;
Is there any problem with my thinking, or is it possible to solve the pinyin error correction and get Chinese results?