Sam,
I'm not an expert, but I think making `co-occurrences` recursive is
slightly simpler. Something like:
(defn co-occurrences [db ht depth tags]
(if (zero? depth)
>
tags
> (recur db ht (dec depth) (co-ocs db ht tags
You'll note that I'm not reducing into tags on the last line. My version of
`co-ocs` returns all of the tags you care about, which also helps simplify
`co-occurrences`.
I also formatted the `co-ocs` function with a thread-last macro to make it
easier to read and manipulate.
I had to add a filter step to `co-ocs` in order for this to work the way
you describe; otherwise, it just returns all of the hashtags I've ever
used. Let me know if you'd like to see that code.
Chris
On Wed, Dec 3, 2014 at 9:49 PM, Sam Raker wrote:
> EDIT
>
> On Wednesday, December 3, 2014 10:45:33 PM UTC-5, Sam Raker wrote:
>>
>> I've got a decent-sized corpus of tweets, organized by hashtag, in a
>> CouchDB db. I'm doing some initial explorations of my data, and was curious
>> about which hashtags show up together in tweets. I want to do a NSA-style
>> "hops" kind of algorithm--get all the hashtags that show up in the same
>> tweets as hashtags that show up in the same tweets as hashtags that show up
>> in the same tweets as my "target hashtag", to an arbitrary depth. I wrote
>> this:
>>
>> (defn co-ocs [db ht & [s]]
>> (reduce into (or s #{})
>> (map #(map :text %)
>> (map #(get-in % [:entities
>> :hashtags])
>> (:tweets (clutch/get-document db ht))
>>
>> (defn co-occurrences [db ht depth]
>> (loop [tags (co-ocs db ht) i 1]
>> (if (<= i depth) (recur
>> (reduce into tags
>> (map (partial co-ocs db) tags))
>> (inc i))
>> tags)))
>>
>> It works, but loop + incrementing a counter seems profoundly un-clojuric.
>> I suppose I could use `dotimes` + an atom, but that doesn't seem much
>> better. Any suggestions?
>>
> --
> You received this message because you are subscribed to the Google
> Groups "Clojure" group.
> To post to this group, send email to clojure@googlegroups.com
> Note that posts from new members are moderated - please be patient with
> your first post.
> To unsubscribe from this group, send email to
> clojure+unsubscr...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/clojure?hl=en
> ---
> You received this message because you are subscribed to the Google Groups
> "Clojure" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to clojure+unsubscr...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
---
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to clojure+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
