Fragrance Dataset for 50K perfumes and colognes

3,000.00

Fragrance Dataset for 50K perfumes and colognes (includes accords, notes, perfume pyramid, silage, longevity and user votes)

Add to cart

Description

The fragrance data is split over 6 comma seperated values (CSV) files.

fragrances.csv – 5 columns:

  1. Fragrance ID: Unique Fragrance ID for each fragrance.
  2. Name: Fragrance name. ‘,’ characters are removed.
  3. Brand: Fragrance brand.
  4. Gender: The gender for which the fragrance was designed for.
  5. Year: The year the fragrance was released. ‘N/A’ indicates that the year was not available.

ratings.csv – 11 columns:

  1. Fragrance ID: Unique Fragrance ID for each fragrance.
  2. Love: Number of users that voted they love this fragrance.
  3. Like: Number of users that voted they like this fragrance.
  4. Dislike: Number of users that voted they dislike this fragrance
  5. Winter: Number of users that voted this is a winter fragrance.
  6. Spring: Number of users that voted this is a winter fragrance.
  7. Summer: Number of users that voted this is a winter fragrance.
  8. Autumn: Number of users that voted this is a winter fragrance.
  9. Day: Number of users that voted this is a winter fragrance.
  10. Night: Number of users that voted this is a winter fragrance.
  11. Total votes: The total number of users that voted. Users can vote for any combination of variables 4-10 and one of variables 1-3. All variables are optional. They can also remove their vote for one or more variables at a later stage, and it still counts as a vote. For this reason this number is *not* equal to the sum of all columns.

accords.csv – 3 columns:

  1. Fragrance ID: Unique Fragrance ID for each fragrance.
  2. Accord: The name of the accord.
  3. Weight: A number between 0 and 1 representing how strong the presence of the accord is.

notes.csv – 4 columns:

  1. Fragrance ID: Unique Fragrance ID for each fragrance.
  2. Note: The name of the note
  3. Type: The type of the note: ‘top‘, ‘middle‘, or ‘bottom‘. ‘N/A‘ indicates that no classification is available.
  4. Votes: The number of users that voted that this note is present in the fragrance.

longevity.csv – 6 columns:

  1. Fragrance ID: Unique Fragrance ID for each fragrance.
  2. Poor: The number of users that voted that the frangrances longevity is poor (less than 1 hour).
  3. Weak: The number of users that voted that the frangrances longevity is weak (1 – 2 hours).
  4. Moderate: The number of users that voted that the frangrances longevity is moderate (3 – 6 hours).
  5. Long lasting: The number of users that voted that the frangrances longevity is long lasting (7 – 12 hours).
  6. Very long lasting: The number of users that voted that the frangrances longevity is very long lasting (over 12 hours).

silage.csv – 5 columns:

  1. Fragrance ID: Unique Fragrance ID for each fragrance.
  2. Soft: The number of users that voted that the frangrances silage is soft (sits close to skin without a trail).
  3. Moderate: The number of users that voted that the frangrances silage is moderate (radiates within arm length).
  4. Heavy: The number of users that voted that the frangrances silage is heavy (radiates within 6 feet).
  5. Enormous: The number of users that voted that the frangrances silage is enormous (fills a room).

Additional information

License

Number of variables (columns)

34

Number of observations (rows)

49408

Reviews

There are no reviews yet.

Be the first to review “Fragrance Dataset for 50K perfumes and colognes”

Your email address will not be published. Required fields are marked *

( 1 rating ) View All Ratings

No ratings have been submitted for this product yet.