Extracción de respuestas

Introducción a la IA Generativa en Snowflake

James Cha-Earley

Senior Developer Advocate, Snowflake

¿Por qué usar extracción de respuestas?

Identificar quejas
Extraer comentarios sobre servicios
Responder preguntas de documentos largos

Un documento con un signo de interrogación rodeado de emojis con caras sonrientes o tristes, representando que no sabemos de qué trata cada reseña

Un problema común de negocios

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 rescate!

Extrayendo una respuesta

from snowflake.cortex import extract_answer


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

Salida

print(response)

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

Puntuaciones de 0 a 1

Construyendo un flujo de trabajo

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

Salida de 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!!!

Haciendo una pregunta

negative_review = cell1.to_pandas()['DESCRIPTION'].iloc[0]


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

print(response)

Revisando la respuesta

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

Limitaciones

Puede devolver frases parciales o vagas
Mejor para párrafos, no documentos de varios temas
Siempre revisa salidas críticas

Señal de precaución

¡Vamos a practicar!

Introducción a la IA Generativa en Snowflake