The purpose of this capstone project is to tie everything we learned in this class together. This might be challenging in the short term, but is consistently rated as being extremely valuable in the long run. The cover story is that you are working as a Data Scientist for an auction house that works with a major art gallery. You want to better understand this space with the help of Data Science. Historically, this domain was dominated by art historians, but is increasingly informed by data. This is where you – a budding Data Scientist – come in. Can you provide the value that justifies your rather high salary?

Mission command preamble: As in general, we won’t tell you *how* to do something. That is up to you and your creative problem solving skills. However, we will pose the questions that you should answer by interrogating the data. Importantly, we do expect you to do this work yourself, so it reflects your intellectual contribution – not that of third parties. By doing this assignment, you certify that it indeed reflects your individual intellectual work.

Format: The  project  consist  of your answers to  10  (equally-weighed, grade-wise)  questions.  Each answer *must* include some text (describing both what you did and what you found, i.e. the answer to the question), a figure that illustrates the findings and some numbers (e.g. test statistics, confidence intervals, p-values or the like). Please save it as a pdf document. This document should be 4-6 pages long (arbitrary font size and margins). About half a page per question is reasonable. In addition, open your document with a brief statement as to how you handled preprocessing (e.g. dimension reduction, data cleaning and data transformations), as this will apply to all answers. Include your name.

Academic integrity: You are expected to do this project by yourself, individually, so that we are able  to determine a grade for you personally. There are enough degrees of freedom (e.g. how to clean the data, what variables to compare, aesthetic choices in the figures, etc.) that no two reports will be alike. We’ll be on the lookout for suspicious similarities, so please refrain from collaborating. To prevent cheating (please don’t do this – it is easily detected), it is very important that you – at the beginning of the code file – seed the random number generator with your N-number. That way, the   correct answers will be keyed to your own solution (as this matters, e.g. for the specific train/test split or bootstrapping). As N-numbers are unique, this will also protect your work from plagiarism.

Failure to seed the RNG in this way will also result in the loss of grade points.

Deliverables: Upload two files to the Brightspace portal by the due date in the sittyba:

*A pdf (the “project report”) that contains your answers to the 10 questions, as well as an introductory paragraph about preprocessing.

*A .py file with the code that performed the data analysis and created the figures.

We do wish you all the  best in executing on these  instructions. We aimed at an optimal  balance between specificity and  implementation  leeway, while still  allowing  us to grade the  projects  in a consistent and fair manner.

Everything should be doable from what was covered in this course.

If you take this project seriously and do a quality job, you can easily use it as an item in your DS portfolio. Former students told us that they secured internships and even jobs by well executed capstone projects that impressed recruiters and interviewers.

Description of dataset: This dataset consists of data from 300 users and 91 art pieces.

The art pieces are described in the file “theArt.csv” . Here is how this file is structured: 1st  row: Headers (e.g. number, title, etc.)

Rows 2-92: Information about the 91 individual art piece

Column 1: “Number” (the ID number of the art piece)

Column 2: Artist

Column 3: Title

Column 4: Artistic style

Column 5: Year of completion

Column 6: Type code – 1 = classical art, 2 = modern art, 3 = non-human art

Column 7: computer or animal code – 0 = human, 1 = computer generated art, 2 = animal art Column 8: Intentionally created? 0 = no, 1 = yes

You can also take a look at the actual art by looking at the files in the “artPieces” folder on Brightspace.

The user data is contained in the file “theData.csv” . Here is how this file is structured: Rows 1-300: Responses from individual users

Columns 1-91: Preference ratings (liking) of the 91 art pieces. The column number in this file           corresponds to the number of the art piece in column 1 of “theArt.csv” file described above. For    instance, ratings of art piece 27 (“the woman at the window”) is in column 27. Numbers represent preference ratings from 1 (“hate it”) to 7 (“love it”).

Columns 92- 182: “Energy” ratings of the same 91 art pieces (in the same order as the preference   ratings above). Numbers represent ratings from 1 (“it calms me down a lot”) to 7 (“it agitates me a lot”).

Columns 183- 194: “Dark” personality traits. Numbers represent how much a user agrees with a statement, from 1 (strongly disagree) to 5 (strongly agree). Here are the 12 statements, in column order:

1           I tend to manipulate others to get my way

2           I have used deceit or lied to get my way

3           I have used flattery to get my way

4           I tend to exploit others towards my own end

5           I tend to lack remorse

6           I tend to be unconcerned with the morality of my actions

7           I can be callous or insensitive

8           I tend to be cynical

9           I tend to want others to admire me

10         I tend to want others to pay attention to me

11         I tend to seek prestige and status

12         I tend to expect favors from others

Columns 195-205: Action preferences. Numbers represent how much a user agrees with a statement, from 1 (strongly disagree) to 5 (strongly agree). Here are the 11 actions, in column order:

1           I like to play board games

2           I like to play role playing (e.g. D&D) games

3           I like to play video games

4           I like to do yoga

5           I like to meditate

6           I like to take walks in the forest

7           I like to take walks on the beach

8           I like to hike

9           I like to ski

10         I like to do paintball

11         I like amusement parks

Columns 206-215: Self-image/self-esteem. Numbers represent how much a user agrees with a statement. Note that if a statement has “reverse polarity”, e.g. statement 2 “at times I feel like I am no good at all”, it has already been re-coded/inverted by the professor such that higher numbers represent higher self-esteem. Here are the 10 items, in column order:

1           On the whole, I am satisfied with myself

2           At times I think I am no good at all

3           I feel that I have a number of good qualities

4           I am able to do things as well as most other people

5           I feel I do not have much to be proud of

6           I certainly feel useless at times

7           I feel that I'm a person of worth, at least on an equal plane with others

8           I wish I could have more respect for myself

9           All in all, I am inclined to feel that I am a failure

10         I take a positive attitude toward myself

Column 216: User age

Column 217: User gender (1 = male, 2 = female, 3 = non-binary)

Column 218: Political orientation (1 = progressive, 2 = liberal, 3 = moderate, 4 = conservative,

5 = libertarian, 6 = independent)

Column 219: Art education (The higher the number, the more: 0 = none, 3 = years of art education) Column 220: General sophistication (The higher, the more: 0 = not going to the opera, etc.

3 = doing everything – opera, art galleries, etc.)

Column 221: Being somewhat of an artist myself? (0 = no, 1 = sort of, 2 = yes, I see myself as an artist)

Note that we did most of the data munging and coding for you already but you still need to handle missing data in some way (e.g. row-wise removal, element-wise removal, imputation) .

Extreme values might also have to be handled.