headline_detector

headline_detector

Indonesian Headline Detection Model Repository

There's a Python library that provides APIs for detecting headlines in textual data, especially on social media platforms such as Twitter. The library utilizes a model that has been developed and trained on a dataset of Twitter posts containing both headline and non-headline texts, with the assistance of journalism professionals to ensure the data quality.

$ pip install headline-detector

Available scenario and the performance

All meassured in accuracy

Model Throughput

Tested on Intel i7-6700k and 32GB of RAM.

Usage

Output either 0 (non-headline) and 1 (headline)

from headline_detector import FasttextDetector, IndoBERTweetDetector, CNNDetector

detector = FasttextDetector.load_from_scenario(1)
data = detector.predict_text(
    [
        "nama kamu siapa?",
        "Kapolda Jatim Teddy Minahasa Dikabarkan Ditangkap Terkait Narkoba  https://t.co/LD9X6VFaUR",
    ]
)
print(data)  # output: [0, 1]

detector = CNNDetector.load_from_scenario(3)
data = detector.predict_text(
    [
        "nama kamu siapa?",
        "Kapolda Jatim Teddy Minahasa Dikabarkan Ditangkap Terkait Narkoba  https://t.co/LD9X6VFaUR",
    ]
)
print(data)  # output: [0, 1]

detector = IndoBERTweetDetector.load_from_scenario(5)
data = detector.predict_text(
    [
        "nama kamu siapa?",
        "Kapolda Jatim Teddy Minahasa Dikabarkan Ditangkap Terkait Narkoba  https://t.co/LD9X6VFaUR",
    ]
)
print(data)  # output: [0, 1]

# 0 is non-headline
# 1 is headline