Estrazione di risposte

Introduzione alla Generative AI in Snowflake

James Cha-Earley

Senior Developer Advocate, Snowflake

Perché usare l'estrazione di risposte?

Individuare i reclami
Estrarre feedback sui servizi
Rispondere a domande da documenti lunghi

Un documento con un punto interrogativo circondato da emoji sorridenti o tristi, a indicare che non sappiamo di cosa parla ogni recensione

Un problema aziendale comune

review = """
The property needs updating and management needs to hire more staff. 
The staff does everything they can but it is not enough for the number of guests. 
The rooms look old as well as everything in the bathrooms. 
Definitely better options in the hotel zone.
"""

Snowflake Cortex al salvataggio!

Estrarre una risposta

from snowflake.cortex import extract_answer


response = extract_answer(
    from_text=review,
    question="What did the guest complain about?"
)

Risultato

print(response)

[
    {
    "answer": "Management needs to hire more staff.",
    "score": 0.8013839254
    }
]

Punteggi da 0 a 1

Creare un workflow

SELECT DESCRIPTION
FROM HOTELS.REVIEWS
WHERE RATING < 6
    AND DATE = '2023-08-23';

Output SQL

I will never book a vacation through Expedia not only did I pay over 5;000 dollars
but they false advertise the resort; in my reservation It clearly stated that
breakfast and dinner was included. When I got to the resort they informed me that 
these issues have happed before with Expedia because the resort does not offer an
all exclusive. After flighting practically all day I called Expedia and they had me
on hold for 45min and nothing got resolved. I ended up spending over 1000.00 more
just on food. I am beyond me they are not only false advertising but refused to
take any responsibility on their part for the issues they caused me and my family.  Again I will never chose Expedia to plan a vacation!!!

Porre una domanda

negative_review = dataframe_1.to_pandas()['DESCRIPTION'].iloc[0]


response = extract_answer(
    from_text=negative_review,
    question="What was the primary issue that the guest mentioned?"
)

print(response)

Verificare la risposta

[
    {
    "answer" : "Expedia falsely advertised that meals were included, but the
                resort did not offer an all-inclusive package.",
    "score": 0.75829312442
    }
]

Limiti

Può restituire frasi parziali o vaghe
Meglio su paragrafi, non su documenti multi-tema
Controlla sempre gli output critici

Segnale di cautela

Esercitiamoci!

Introduzione alla Generative AI in Snowflake