Semantle: Can you think like a computer?

got today's in 30, made a leap from mechanic to the secret word

Yeah my son beat my score today by a few guesses, he came to the answer by a different route though. my guesses went from star to celebrity to leader and then other notable types of people while he found it with the multiple body parts to anatomy to doctor etc.

I gave up on yesterdays because my 4 closest guesses were (iirc) PERHAPS, NEITHER, uhh maybe SILKILY??!, and EITHER. None of them were going to get me to the answer, in retrospect.

*cracks fingers* let's try today's.

Today's in 64. It took me a while to start scoring, and then when I did all mine were eye related, so I was in a similar ballpark but far too specific!

I did give up yesterday but only because I had to go to bed after having the page up all day to try new things - yes, I am one of those people who took "it's too hard" as a challenge. I got green scores at 25, 990 at 54, and then pages upon pages of green, very few of which were related to the answer in a way I would understand. My highest match was IRONICALLY, ffs. I was working on a few different strands, one was conjunctions and old-timey constructions (I had a lot of fun with inasmuches and heretofores), one was a sort of "quirky" adverb strand (QUIXOTICALLY and NEBULOUSLY and BIZARRELY all scored green), and I got a few that could theoretically have led me to the answer (SUBSEQUENTLY and CONSEQUENTLY), however more obvious matches like AFTERWARDS and LASTLY were cold, so I wasn't looking in that direction.

Honestly, I love this game despite the frustration at machine-calculated relations.

I had pretty much the same strands, emil.y! Less exhaustively than yours, I imagine, tho I even ran through all the contractions thinking maybe that was what made it "difficult."

I mean...SILKILY???! lol it's inexplicable.

Got it in 144! Possibly my best day yet tho I maybe took a funny route.

Got it in 25 today, which is my best yet.

Took me ages but I got there. Highest was tendon] for a long time, then cardiac which led me to implant and scalpel

I have no idea how anyone does these in less than 100. I think I kind of enjoy typing random words into a computer text box... as my ILX posting history will demonstrate!!!!!!!!!!!!!!!!!!!!

lol otm

I looked back on the "tail" one from the other day because I still had it open in a tab, and noted that while I got there from typing in "bird" then "flapping" then "wing" then various aeroplane/helicopter-related terms (it liked "propeller" and "rotor" quite a lot) and then typed "tail" while I was considering helicopter parts.

(I haven't spoiler-tagged these since it's not today's puzzle; sorry if anyone feels I should have done!)

Meanwhile, the names of various animals which have tails and particular noted behaviours thereof (dog, cat, pig) scored way less than bird, as did other things with tails (shirt and, uh, sperm), and things like "follow" and "stalk" which are synonyms of "tail" in some contexts scored cold too.

It's a mystery. But it's a fun mystery. Also kinder otm, such fun to type random words and follow weird tangents while imagining "the computer must be thinking really hard about why I chose this one!!"

xxp

I wouldn't feel bad at being crap at it, because you definitely aren't alone! But I feel like once you start getting a feel of how the word2vec data set works, then it will become even harder!

Kinder--I took a totally different route! My first hits in the green were BARBER and BEAUTICIAN, which then led (eventually) to CLINICIAN, and then I was in the home stretch. Random!

Just got the new one! 12 minutes, 108 guesses, but this is definitely a speed record for me.

I solved Semantle #18 in 133 guesses. Again, felt like a pretty linear process (rewardingly so) once I got the first green bar!

I solved Semantle #18 in 159 guesses. Went on a weird tangent when I got a hit on online, and then pay in the top 100.

got it in 127. sadly took me 70 tries after I got my first green to realize I'd been going the opposite path that it wanted.

benefits was my first word approximating 20% similarity, and then pension was the first that got me green. all of my work jargon was helpful this time around.

53, my best yet - got lucky with a green on my 9th guess

111, not too bad a showing. If I’d followed an instinct earlier I probably could have halved that, but I spent a long time chasing a dead end.

Meanwhile, the names of various animals which have tails and particular noted behaviours thereof (dog, cat, pig) scored way less than bird

I'm guessing this is because 'bird' is a larger class than 'dog', 'pig', etc, and any text about parts of a bird or a particular species will use the same other words (beak, wings) whereas if it's about a cat it will mention paws and whiskers, if it's a pig it will mention trotters and snout, etc. 'Nose' was pretty high up though.

From memory nose, snout and beak and maybe proboscis all scored pretty highly on that one - which had me totally stuck looking for “animal face parts” - didn’t think for a second to look at the other end!

Got today's in 162 because I am an idiot.

I got month in 65! Why did it take me another 100 guesses?

Stuck on the third closest word. No clue.

ha my 13th word just hit 989...total luck

I tried this for the first time and rage quit / gave up after 45 attempts all below 12.

but now it's got interesting, i've had four guesses since and not got higher than that one. From the info it seems my best guess is 10th. And I'm absolutely stumped. and scared. because the perfectionist in me wants to guess this right on my next guess.

this is my first time in the zone. so regular zone visitors, how many guesses on average do you usually take once you're very close?

The data set is small at the moment but, usually, loads.

Lol I did it for the first time yesterday and managed to get really close without having read the instructions and I thought similarity meant word composition ie combinations of letters rather than meaning

Weirdly I was able to make pretty good progress under this misapprehension

omg i got fucking burnt on this !!! haha

87! I could have done it in fewer if I'd realised that I hadn't already tried the answer - I had daily, hourly, weekly, yearly, so I just assumed I'd tried the answer already, doh. (That is particularly spoilery hidden text, don't click if you haven't got it yet.)

regular zone visitors, how many guesses on average do you usually take once you're very close?

Bloody loads, honestly. But I get addicted to the dopamine hit of the green. Can't stop playing. Never stop playing. Give me that green score baby.

Does anyone else resort to typing words in the intro blurb if their first few random guesses don't get anywhere near? It actually seems to be quite a productive method of initially guessing. I also do some previous answers as I know what word2vec thinks is close to them now.

haha yes

well 31 guesses, but I got sooo lucky, and then effed up, could have been less. I won't say anything, feared that I will mess up the spoiler text formatting tbf. I will reveal in a few days.

Definitely done that, yes! Or played while half-watching TV and typed words from the TV in.

Once I'm quite close, it takes me anywhere between a few guesses to several hundred guesses. On average? No idea but probably tens. The mean is probably 100+ after a couple of outliers (including the one flagged with a warning) where I didn't get it for ages after hitting the 990s.

this game is fun!

219 for me. Did that thing of being completely stuck and out of guesses despite being on 977. Went back to it later and it only took a few more.

I got up to 977/1000 and got fucked off + gave up and damn it I should have spent at least another half an hour, because was so nearly there. But it's all good experience for getting a feel for how this word2vec scores the word similarity. If I ever see anyone claiming today's wordle was hard I'm going to link this to them from now on, it's child's play in comparison.

I managed it in 31 today. Seems like low similarity scores (<20) are words that might appear in the same sentence as the target word (how big of a content window is the algorithm using?), medium scores (21-40) are words that commonly appear adjacent to the target word, and medium/high scores (>40) start trending towards actual synonyms or members of the same semantic category?

sorry, context window not content

63 in the end

Took 43 to get a green

Solved the new semantle, #19, in 2 minutes, 36 guesses

got it in 61. trimming the fat a little, but once again, I could have saved ten guesses if I pivoted earlier once I knew I was heading in the right direction.

I solved Semantle #19 in 41 guesses. Most straightforward experience yet, still fun. First green was something crazy high like #969.

I have now three other friends playing this

26. far and away my best result, and looking at others' results, this seems a particularly easy one.