GerSent

GerSent is a project that tries to make a relativ good sentiment of a german sentence. Our first approach will be simple techniques like wordlist or interpration to assess the given sentence.

Getting Started

First you have to install our package.

pip install gersent

Now you are able to run a simple sentiment analysis over a german sentence.

from gersent import GerSent

ger_senti = GerSent()

print(ger_senti.sentiment('Ich bin ein guter Satz.')

Configurations

Examples

Positiv

# Die Äpfel schmecken gut.
{'negative': 0.0, 'positive': 0.3716, 'composite': 0.3716}

# Die Äpfel schmecken gut!
{'negative': 0.0, 'positive': 0.39018, 'composite': 0.39018}

# Die Äpfel schmecken gut!!!
{'negative': 0.0, 'positive': 0.42733999999999994, 'composite': 0.42733999999999994}

# Die Äpfel schmecken sehr gut!
{'negative': 0.0, 'positive': 0.39018, 'composite': 0.39018}

Negativ

# Die Äpfel schmecken nicht gut.
{'negative': -0.0, 'positive': -0.3716, 'composite': -0.3716}

References

Wordlists

R. Remus, U. Quasthoff & G. Heyer: SentiWS - a Publicly Available German-language Resource for Sentiment Analysis. In: Proceedings of the 7th International Language Ressources and Evaluation (LREC'10), 2010