Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Funny item co-occurrences in 3.2M Instacart orders

rogerdickey.com

Post reply on HN
Print view View original

41–50 of 103 posts

Re: Funny item co-occurrences in 3.2M Instacart orders

#41

Post by lukewarm707 »

embedding-shape wrote: post #2

> Much better! Some of these even made me laugh out loud. Kale and an enema? Parsley and condoms? Adult diapers and baby food? > Small orders are less common but we still got some fun ones. Oreos and lube? Sounds like a good time! Funny to who ? Was this rated "for sure funny" by a LLM or what's going on? Why is it funny to buy Oreo and lube? I could understand "contradictions" or something like that (like buying wei…

i enjoyed the author sense of play and enthusiasm.

with play they surely gained much domain understanding and source of new ideas.

bad for company and society to enforce the oppressive conformity.

Re: Funny item co-occurrences in 3.2M Instacart orders

#42

Post by StilesCrisis »

anonu wrote: post #6
disillusioned wrote: post #3

Time for me to re-post my perennial "fun banana facts" post: Bananas are the #1 most-sold item at most grocery stores including, notably, Wal-Mart. Bananas also have the highest standard deviation in terms of predicting if a given (known) consumer will purchase bananas in a given store run. (At least as compared to other food products and consumables.) When predicting a consumer's shop, it's generally pretty easy to…

Berries are most certainly not seasonal anymore. They should be but we thoroughly engineered the seasonality out of them. They're always on the shelf. Do people's purchase habits follow the natural seasonality of the product anymore? Probably something that can be found in this dataset as well.

We absolutely buy berries (and tomatoes) when in season. The flavor is much better!

Re: Funny item co-occurrences in 3.2M Instacart orders

#43

Post by 01284a7e »

anonu wrote: post #6
disillusioned wrote: post #3

Time for me to re-post my perennial "fun banana facts" post: Bananas are the #1 most-sold item at most grocery stores including, notably, Wal-Mart. Bananas also have the highest standard deviation in terms of predicting if a given (known) consumer will purchase bananas in a given store run. (At least as compared to other food products and consumables.) When predicting a consumer's shop, it's generally pretty easy to…

Berries are most certainly not seasonal anymore. They should be but we thoroughly engineered the seasonality out of them. They're always on the shelf. Do people's purchase habits follow the natural seasonality of the product anymore? Probably something that can be found in this dataset as well.

They are highly seasonal. We have not "thoroughly engineered the seasonality out of them".

Re: Funny item co-occurrences in 3.2M Instacart orders

#45

Post by embedding-shape »

trebligdivad wrote: post #38
disillusioned wrote: post #3

Time for me to re-post my perennial "fun banana facts" post: Bananas are the #1 most-sold item at most grocery stores including, notably, Wal-Mart. Bananas also have the highest standard deviation in terms of predicting if a given (known) consumer will purchase bananas in a given store run. (At least as compared to other food products and consumables.) When predicting a consumer's shop, it's generally pretty easy to…

The initial grouping of 'pack of organic bananas' with individual bananas feels like a wrong number in the pack problem; eg we typically want one a day so want to buy 7 in a weekly shop but the pack size is 5.

> eg we typically want one a day so want to buy 7 in a weekly shop but the pack size is 5.

Can't you just grab a bigger cluster/bunch and remove N bananas so it has the amount you want? Or remove 3 from the bunch that has five so two 5+2 clusters? Feels like I'm missing something obvious here.

Re: Funny item co-occurrences in 3.2M Instacart orders

#46

Post by LourensT »

embedding-shape wrote: post #45
trebligdivad wrote: post #38

Earlier quoted context omitted.

The initial grouping of 'pack of organic bananas' with individual bananas feels like a wrong number in the pack problem; eg we typically want one a day so want to buy 7 in a weekly shop but the pack size is 5.

> eg we typically want one a day so want to buy 7 in a weekly shop but the pack size is 5. Can't you just grab a bigger cluster/bunch and remove N bananas so it has the amount you want? Or remove 3 from the bunch that has five so two 5+2 clusters? Feels like I'm missing something obvious here.

Yes, because Instacart is ordering for delivery :)

Re: Funny item co-occurrences in 3.2M Instacart orders

#47

Post by embedding-shape »

LourensT wrote: post #46
embedding-shape wrote: post #45

Earlier quoted context omitted.

> eg we typically want one a day so want to buy 7 in a weekly shop but the pack size is 5. Can't you just grab a bigger cluster/bunch and remove N bananas so it has the amount you want? Or remove 3 from the bunch that has five so two 5+2 clusters? Feels like I'm missing something obvious here.

Yes, because Instacart is ordering for delivery :)

Oh yeah, of course, guess I kind of forgot the overall context, and I was thinking in-person shopping... Thanks! :)

Re: Funny item co-occurrences in 3.2M Instacart orders

#48

Post by manwe150 »

amelius wrote: post #16

My conclusion: there's nothing funny about groceries, no matter how you order them. Counterexamples welcome. (Makes sense as I never felt the urge to laugh after looking in someone else's shopping basket. Anyway, perhaps that's why I'm not a data analyst.)

A few I’ve seen in person are someone buying about a dozen pallets of water and a half dozen of a vegetable (it was cabbage or leek if I recall). But that was more about the absurd looking quantity while only wanting just those two items. I assume most of the “funny orders” I see are restaurant owners who already got most of their items delivered, which makes them a large deviation from the expected family purchaser. This is something I have only seen at specific stores (similar to costco) in specific cities.

Re: Funny item co-occurrences in 3.2M Instacart orders

#49

Post by xg15 »

disillusioned wrote: post #3

Time for me to re-post my perennial "fun banana facts" post: Bananas are the #1 most-sold item at most grocery stores including, notably, Wal-Mart. Bananas also have the highest standard deviation in terms of predicting if a given (known) consumer will purchase bananas in a given store run. (At least as compared to other food products and consumables.) When predicting a consumer's shop, it's generally pretty easy to…

I imagine people either buy single bananas as a quick snack or bananas in bulk for breakfast. (Ignoring the "funny" uses here). I wonder if this kind of irregular pattern is a reflection on people's breakfast habits.
Post reply on HN