# Spam_detect function optional setting
## Parameter intro
- accout detect (bool, optional, defaults to True) - adopt detect account & phone message, default not adopt
- url_detect (str, optional, defaults to None)
- None : disable detect
- 're' : only detect url by regular expression (lower accuracy)
- 'urlextract' : detect url by urlextract (higher accuracy)
- language_classify (bool, optional, defaults to True) - adopt language classify, default adopt
- rulebase_detect (bool, optional, defaults to True) - adopt rulebase detection, default adopt
- model_predict (bool, optional, defaults to True) - adopt TextCNN model prediction, default adopt
- pinyin_mode (bool, optional, defaults to False) - adopt pinyin fix module, default not adopt
## analysis
### Full function
#### account_detect=True, url_detect='urlextract', language_classify=True, rulebase_detect=True, model_predict=True, pinyin_mode=True
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(message, account_detect=True, url_detect='urlextract', language_classify=True, rulebase_detect=True, model_predict=True, pinyin_mode=True)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [0, 0, 0, 0, 0, 0, 0],
'ban': True,
'final_label': 5,
'url_count': 0,
'url_list': [],
'account_count': 1,
'account_list': ['SH9709200'],
'language': 'zh',
'predict_result': [],
'status': True
}
--------------in predetect.txt--------------
加微SH9709200 价格:7元105万/ 40元1250万/ -> ["拼音: ([['jia4'], ['ge2']]), Label: 5"]
```
### account_detect
#### account_detect=True, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(message, account_detect=True, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [0, 0, 0, 0, 0, 0, 0],
'ban': True,
'final_label': 5,
'account_count': 1,
'account_list': ['SH9709200'],
'status': True
}
```
### url_detect
#### account_detect=False, url_detect='urlextract', language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(message, account_detect=False, url_detect='urlextract', language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [0, 0, 0, 0, 0, 0, 0],
'ban': True,
'final_label': 5,
'url_count': 0,
'url_list': [],
'status': True
}
```
### language_classify
#### account_detect=False, url_detect=None, language_classify=True, rulebase_detect=False, model_predict=False, pinyin_mode=False
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(message, account_detect=False, url_detect=None, language_classify=True, rulebase_detect=False, model_predict=False, pinyin_mode=False)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [0, 0, 0, 0, 0, 0, 0],
'ban': True,
'final_label': 5,
'language': 'zh',
'status': True
}
```
### rulebase_detect
#### account_detect=False, url_detect=None, language_classify=False, rulebase_detect=True, model_predict=False, pinyin_mode=False
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(message, account_detect=False, url_detect=None, language_classify=False, rulebase_detect=True, model_predict=False, pinyin_mode=False)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [0, 0, 0, 0, 0, 0, 0],
'ban': True,
'final_label': 5,
'status': True
}
```
### model_predict
#### account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=True, pinyin_mode=False
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(message, account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=True, pinyin_mode=False)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [90.91, 0.0, 0.0, 0.0, 0.0, 9.09, 0.0],
'ban': True,
'final_label': 5,
'predict_result': [('加微', 0), ('SH9709200', 0), ('价格', 5), ('7', 0), ('元', 0), ('105', 0), ('万', 0), ('40', 0), ('元', 0), ('1250', 0), ('万', 0)],
'status': True
}
```
### pinyin_mode
#### account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=True
```=python
message = '加微SH9709200 价格:7元105万/ 40元1250万/'
result = model.analysis(account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=True)
print(result)
--------------result--------------
{
'message': '加微SH9709200 价格:7元105万/ 40元1250万/',
'label_distribution': [0, 0, 0, 0, 0, 0, 0],
'ban': True,
'final_label': 5,
'status': True
}
--------------in predetect.txt--------------
加微SH9709200 价格:7元105万/ 40元1250万/ -> ["拼音: ([['jia4'], ['ge2']]), Label: 5"]
```
## batch_analysis
### Full function
#### account_detect=True, url_detect='urlextract', language_classify=True, rulebase_detect=True, model_predict=True, pinyin_mode=True
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=True, url_detect='urlextract', language_classify=True, rulebase_detect=True, model_predict=True, pinyin_mode=True)
print(message_df)
--------------result--------------
message label_distribution ban final_label url_count url_list account_count account_list language predict_result
0 他妈的 [0, 0, 1, 0, 0, 0, 0] True 2 0 [] 0 zh [(他妈的, 2)]
1 台湾加油 [1, 1, 0, 0, 0, 0, 0] True 1 0 [] 0 zh [(台湾, 1), (加油, 0)]
2 加微：SH9709200 [0, 0, 0, 0, 0, 0, 0] True 5 0 [] 1 [SH9709200] zh []
3 http://reurl.cc/6666 [0, 0, 0, 0, 0, 0, 0] True 5 1 [http://reurl.cc/6666] 0 fr []
}
--------------in predetect.txt--------------
他妈的 -> ["拼音: ([['ta1'], ['ma1']]), Label: 2", "拼音: ([['ma1'], ['de']]), Label: 2", "拼音: ([['ta1'], ['ma1'], ['de']]), Label: 2"]
台湾加油 -> ["拼音: ([['tai2'], ['wan1']]), Label: 1"]
```
### account_detect
#### account_detect=True, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=True, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False)
print(message_df)
--------------result--------------
message label_distribution ban final_label account_count account_list
0 他妈的 [0, 0, 0, 0, 0, 0, 0] False 0 0
1 台湾加油 [0, 0, 0, 0, 0, 0, 0] False 0 0
2 加微：SH9709200 [0, 0, 0, 0, 0, 0, 0] True 5 1 [SH9709200]
3 http://reurl.cc/6666 [0, 0, 0, 0, 0, 0, 0] False 0 0
```
### url_detect
#### account_detect=False, url_detect='urlextract', language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=False, url_detect='urlextract', language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=False)
print(message_df)
--------------result--------------
message label_distribution ban final_label url_count url_list
0 他妈的 [0, 0, 0, 0, 0, 0, 0] False 0 0 []
1 台湾加油 [0, 0, 0, 0, 0, 0, 0] False 0 0 []
2 加微：SH9709200 [0, 0, 0, 0, 0, 0, 0] False 0 0 []
3 http://reurl.cc/6666 [0, 0, 0, 0, 0, 0, 0] True 5 1 [http://reurl.cc/6666]
```
### language_classify
#### account_detect=False, url_detect=None, language_classify=True, rulebase_detect=False, model_predict=False, pinyin_mode=False
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=False, url_detect=None, language_classify=True, rulebase_detect=False, model_predict=False, pinyin_mode=False)
print(message_df)
--------------result--------------
message label_distribution ban final_label language
0 他妈的 [0, 0, 0, 0, 0, 0, 0] False 0 zh
1 台湾加油 [0, 0, 0, 0, 0, 0, 0] False 0 zh
2 加微：SH9709200 [0, 0, 0, 0, 0, 0, 0] False 0 zh
3 http://reurl.cc/6666 [0, 0, 0, 0, 0, 0, 0] False 0 fr
```
### rulebase_detect
#### account_detect=False, url_detect=None, language_classify=False, rulebase_detect=True, model_predict=False, pinyin_mode=False
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=False, url_detect=None, language_classify=False, rulebase_detect=True, model_predict=False, pinyin_mode=False)
print(message_df)
--------------result--------------
message label_distribution ban final_label
0 他妈的 [0, 0, 0, 0, 0, 0, 0] False 0
1 台湾加油 [0, 0, 0, 0, 0, 0, 0] False 0
2 加微：SH9709200 [0, 0, 0, 0, 0, 0, 0] False 0
3 http://reurl.cc/6666 [0, 0, 0, 0, 0, 0, 0] False 0
```
### model_predict
#### account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=True, pinyin_mode=False
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=True, pinyin_mode=False)
print(message_df)
--------------result--------------
message label_distribution ban final_label predict_result
0 他妈的 [0, 0, 1, 0, 0, 0, 0] True 2 [(他妈的, 2)]
1 台湾加油 [1, 1, 0, 0, 0, 0, 0] True 1 [(台湾, 1), (加油, 0)]
2 加微：SH9709200 [2, 0, 0, 0, 0, 0, 0] False 0 [(加微, 0), (SH9709200, 0)]
3 http://reurl.cc/6666 [5, 0, 0, 0, 0, 0, 0] False 0 [(http, 0), (re, 0), (url, 0), (cc, 0), (6666,...
```
### pinyin_mode
#### account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=True
```=python
messages = ['他媽的','台灣加油','加微：SH9709200','http://reurl.cc/6666']
message_df = model.batch_analysis(messages, account_detect=False, url_detect=None, language_classify=False, rulebase_detect=False, model_predict=False, pinyin_mode=True)
print(message_df)
--------------result--------------
message label_distribution ban final_label
0 他妈的 [0, 0, 0, 0, 0, 0, 0] False 0
1 台湾加油 [0, 0, 0, 0, 0, 0, 0] False 0
2 加微：SH9709200 [0, 0, 0, 0, 0, 0, 0] False 0
3 http://reurl.cc/6666 [0, 0, 0, 0, 0, 0, 0] False 0
--------------in predetect.txt--------------
他妈的 -> ["拼音: ([['ta1'], ['ma1']]), Label: 2", "拼音: ([['ma1'], ['de']]), Label: 2", "拼音: ([['ta1'], ['ma1'], ['de']]), Label: 2"]
台湾加油 -> ["拼音: ([['tai2'], ['wan1']]), Label: 1"]
```
## Precautions
:::warning
1. model_predict 開啟時，若 account_detect or url_detect 也開啟，不會做模型的預測，此訊息會直接歸類至廣告(5)
2. pinyin_mode 需依賴 model_predict，才會做切詞預測後的拼音修正，單獨開啟只會做拼音的預偵測，其結果會例外存在 predetect.txt 裡
3. account_detect & url_detect 兩者皆採用即為廣告偵測的開關
:::
Sign in with Wallet
Connect another wallet