Added double sides :) for DOUBLE accuracy. Found a bug where every word got often a '' with no word, big oops !
Added minimum = 10 counts like the others do. It warms you this is Not going to do well for accuracy. Made all caps lowercased. 1/6th or so More data now! Fixed the algorithm so it can't screw up just in case. Also I found a big problem. Common words....I was looking IN the list of contexts, not globally enwik9.txt! So hockey=247 would get it and the=127 a bit less, when should be otherrr way around!! the=4667246, hockey=452 essentially. Calibrated the in-list top freq restraint. lookin really good now ! Also I'm getting this feeling like they are cheating lol like as if they are using nearby words as the related instead of as proof, I mean try in their link above 'computer'......really? 'delete' and 'damage' is related word? Maybe topical to prime but not to actually translate to directly on the spot. Nonetheless, I am not yet doing this related priming lol. Upper code is just tree stuff (and, dataset searcher cuz i was too lazy to do tree for that part), it looks confusing sorry lol! Bottom is the list comparer! Words I decided to compare to ICE and COMPUTER (I tried some for both ex. ice=machine?): ICE? snow=0.11322215741147319 cold=0.09415304391949597 her=0.07750462707068446 frozen=0.07259386871900203 wind=0.06986131470667552 country=0.06866124047133551 data=0.0645871971970835 solid=0.06261448184752398 rocks=0.061390092996170036 house=0.05842006036774735 ran=0.05690442407640973 cars=0.05625748000083986 planet=0.05620970244457469 runs=0.056124227880248775 girl=0.051638288508951295 drive=0.05115981747258963 glacier=0.05055836507585705 look=0.047316799815272176 woman=0.04572853445553063 crust=0.04283590121218987 ghost=0.0410167500080067 machine=0.040883908295542525 mother=0.04049203529605726 wet=0.04027958222957126 melted=0.03872938548932658 motor=0.03864754595853467 software=0.0380583676431136 computer=0.037836190751224186 moved=0.03551215128396339 jump=0.03371481037000622 virus=0.03369885239775037 iced=0.033190532013929595 equipment=0.03246905924605326 codes=0.028123422968783784 processor=0.026744615916900343 happy=0.02545016681442651 freeze=0.02211497724137182 crystals=0.018834258524980173 delete=0.01506740681998414 raced=0.01288659793814433 rode=0.011895321173671689 aunt=0.01155900120109418 flakes=0.010309278350515464 cute=0.007335448057097542 snows=0.0055511498810467885 exited=0.0049563838223632035 filesystems=0.0017842981760507533 leaped=0.0013877874702616971 COMPUTER? software=0.2459529607379434 data=0.10863912932059587 food=0.09314391618522774 system=0.09214996731090673 workstation=0.08996734010779821 machine=0.08289191038358397 processor=0.07328534254644654 book=0.06885126154442518 vehicle=0.06391849258038963 equipment=0.06092936199605734 her=0.05948213538545814 could=0.055938417882751564 country=0.055922503225975244 house=0.05457888129428688 after=0.05030710594051818 device=0.05011226027097931 brain=0.04831150343586642 truck=0.047897439575114845 table=0.045383152221026325 bathroom=0.04361508308949373 website=0.038618616832237634 dog=0.03843687958021132 ice=0.03783619075122418 bird=0.036951109262121105 wind=0.035948801255487065 love=0.03561035863609414 flying=0.03356048147334977 cold=0.033147939257507 planet=0.03168334537528895 virus=0.030077051409700663 moon=0.029999859849798707 damage=0.029257813653811325 solid=0.028415578442110387 woman=0.027129976538562355 junk=0.02564883443546052 river=0.02487387235335279 liquid=0.024723840606259907 cook=0.02413970976254965 bed=0.024005750344440892 snow=0.022538512816464865 walking=0.02050292203122577 rocks=0.01984966418696772 cute=0.01930955491933795 insect=0.01877782252451295 rain=0.01835271820295515 seller=0.01615162512918414 wet=0.015500301209636577 shirt=0.015256542128461078 frozen=0.013651280569497054 aunt=0.01187750352063693 ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T01fa5e447808d368-M928766a7486c0ff9e748d4c5 Delivery options: https://agi.topicbox.com/groups/agi/subscription
