Datasets:
seoulalpha
/
dataset

Dataset card Data Studio Files
xet
Community
1
text
stringlengths
0
624
food_percent, shopping_percent (비율: 0~100)
만족도 관련 변수 (1~5점)
- food, landscape, heritage, language, safety, budget, accommodation, transport, navigation
- "매우 만족" = 5
- "만족" 또는 "대체로 만족" = 4
- "보통" = 3
- "불만족" 또는 "대체로 불만족" = 2
- "매우 불만족" = 1
Based on the user's sentence, extract the following variables and return them in the specified JSON format.
If a value is not explicitly stated or cannot be inferred, assign it as `null`.
Make sure to return only the JSON object, without any comments, explanations, or extra text.
The key names and order in the JSON output must be preserved as below.
You may assume that the user's input may be in any language (Korean, English, Japanese, etc.).
Respond appropriately by interpreting the input regardless of its language.
[Expected JSON format]
{
"country": (int),
"gender": (int),
"age": (int),
"revisit_indicator": (int),
"visit_local_indicator": (int),
"planned_activity": (int),
"stay_duration": (int),
"accommodation_percent": (float),
"food_percent": (float),
"shopping_percent": (float),
"food": (int),
"landscape": (int),
"heritage": (int),
"language": (int),
"safety": (int),
"budget": (int),
"accommodation": (int),
"transport": (int),
"navigation": (int)
}
[Mapping rules]
gender:
- "male", "man", "남성", "남자" = 1
- "female", "woman", "여성", "여자" = 2
age group:
- "10s", "10대" = 1
- "20s", "20대" = 2
- "30s", "30대" = 3
- "40s", "40대" = 4
- "50s", "50대" = 5
- "60s or older", "60대", "70대" = 6
country region:
- East Asia (China, Japan, Taiwan, etc.) = 1
- Southeast Asia (Thailand, Vietnam, Singapore, etc.) = 2
- Middle East and India = 3
- English-speaking (USA, UK, etc.) = 4
- Europe (France, Germany, etc.) = 5
- Others = 99
revisit_indicator:
- Any phrase indicating re-visit (e.g., "came again", "visited again", "second visit", "return") = 1
- First-time visit (e.g., "first time", "never been before") = 0
visit_local_indicator:
- Mention of non-Seoul locations (e.g., "visited Busan", "went to Jeju", "outside Seoul") = 1
- Only Seoul (e.g., "only visited Seoul", "did not leave Seoul") = 0
planned_activity:
- K-pop, fan meetings, concerts = 1
- Nature, heritage, traditional culture, landscape = 2
- Food, restaurants, local cuisine = 3
- Shopping, duty-free = 4
- Sports, leisure = 5
- Medical, beauty, wellness = 6
- Nightlife, amusement parks = 7
- Trendy, modern culture = 8
- Other = 99
stay_duration:
- Any explicit mention of stay duration such as “3 days”, “4 nights 5 days”, etc. = extract as integer (in days)
accommodation_percent / food_percent / shopping_percent:
- If described (e.g., "half of my spending was for shopping"), estimate a percentage (0–100). Otherwise set to null.
Satisfaction scores (1–5) for:
- food, landscape, heritage, language, safety, budget, accommodation, transport, navigation
→ Map as:
- “very satisfied” = 5
- “satisfied”, “mostly satisfied” = 4
- “neutral”, “average” = 3
- “dissatisfied”, “somewhat dissatisfied” = 2
- “very dissatisfied” = 1