4th March_ proposal of wine research
- caiqingjingyi
- Mar 4, 2020
- 1 min read
Draft:
Source of the data: http://archive.ics.uci.edu/ml/datasets/Wine
Digging into UCI machine learning repository, wine dataset distinguishes from other dataset, winning the top hits among all.
Background information on what the data is about:
The wine dataset contains the results of a chemical analysis of wines grown in a specific area of Italy. Three types of wine are represented in the 178 samples, with the results of 13 chemical analyses recorded for each sample. Notably, it is no longer a fresh dataset, due to its popularity as a benchmark dataset in the field of machine learning, especially classification.
Total Dimensions: 14 with non-missing 178 samples (observations). By original, there exists 13 numerical dimensions, 1 categorical dimension. However,14 dimensions would contribute fairly less if they have been completely taken into account. Yet, at the same time, no external information telling which of the dimensions overwhelm others comes in hand.
Your research questions: general description of the type of insights
How does vinification based on three cultivars across Italy differ from each other in chemical characteristics?
Sounds like an interesting topic, and maybe technically hard for people who are not into wine to understand! I would advise the group to maybe include, besides the regular data exploration planned, some coverage of the terminology involved. After taking a look at the 13 available dimensions, most of them don't seem to be of intuitive understanding for the general public. Good luck!