Metadata-Version: 2.1
Name: autosentiment
Version: 1.1.4
Summary: An automatic sentiment analysis pakage
Home-page: https://github.com/CodeFighter03/autosentiment
Author: Sazin Reshed Samin
Author-email: sazinsamin50@gmail.com
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown

# Automate sentiment analysis tool

### Author : Sazin Reshed Samin

* Email : <sazinsamin50@gmail.com>


## autosentiment is an open source library that generates sentiment type(positive,negetive,neutral) pie char,percentage,number and ternary value for pandas dataframe text portion.


##- Usage
For analysis the seintiment type in positive,negetive or neutral


## - Setup in normal environment and command window:
```
pip install autosentiment
```


## - Setup in jupyter notebook:
```
!pip install autosentiment
```


## - Import library : 
```
import autosentiment as at
```


## - The library is pandas dataframe dependent.
```
Have to get dataframe('text columns') and give to command.
Like df['text]
```




## Features
### - sentiment type pie chart :
```
at.pie()
```

### - sentiment type amount : 
Get the sentiment type(postive,negetive,neutral numbers)
```
at.number()
```


### - sentiment percentage :
Get the percentage of sentiment type
```
at.percentage()
```


### - An example usages

```

>>import autosentiment as at

>>import pandas as pd

>>df=pd.read_csv("/home/samin/anaconda3/dataset_2.csv")

>>percent=at.percentage(df['text'])

>>print(percent)
>>Positve : 33.31 %, Negetive 20.96 %, Neutral : 45.72 %

>>number=at.number(df['text'])

>>print(number)
>>{'positive  ': 1087, 'negetive': 684, 'neutral': 1492}

>>ana=at.analysis_ternary(df['text'])

>>print(ana)
>>[-1, 1, 0.0, 0.0, 0.0, 0.0,.......,1]

>>at.pie(df['text'])


```
![pie chart](../home/samin/Videos/image_12.png)




* For any bug, please notify in my email : <sazinsamin50@gmail.com>







