#ïŒ.ç°¡åãªæŠèŠ
ãã®èšäºã§ã¯éœå
ã©ãŒã¡ã³å±ã®é£ã¹ãã°å£ã³ãã䜿ã£ãŠWord2vecã§ã¢ãã«æ§ç¯ãããããæ¹ã解説ããŠãããŸãã
ç§èªèº«ðã倧奜ãã§æã¯å¹Žé100æ¯ä»¥äžé£ã¹æ©ããŠããèªç§°ã©ãŒã¡ã³ã¬ãå¢ã§ããããããªãããçŽè¿ã®å¥åº·èšºæã«ã²ã£ããããå»è
ãããã¯ã¿ãŒã¹ãããããããããŠããŸããŸããããã
è¡ãå Žããªãããã©ãŒã¡ã³ç±ãçºæ£ãã¹ãæ©æ¢°åŠç¿ã§ã©ãŒã¡ã³ã¬ã³ã¡ã³ãïŒé ããååºãã¬ã³ã¡ã³ãã§çºæïŒã«ææŠããŠã¿ãããšã«ããŸããã
ä»åã¯ãã¹ã¯ã¬ã€ãã³ã°ã§ååŸããå£ã³ããåŠç¿ããŒã¿ãšããŠWord2vecã§ã¢ãã«æ§ç¯ãè¡ããŸãã
äžã®äžã«é£ã¹ãã°å£ã³ãããŒã¿ã§ã¢ããªã³ã°ããèšäºã¯ããã€ããããŸãããã
ã©ãŒã¡ã³å±ã®"ååºã®å£ã³ãã®ã¿"ã§ã¢ãã«æ§ç¯
ããã£ãŠããåäŸãèŠåœãããªãã£ãã®ã§ãã£ãŠã¿ãŸããã
æ¬èšäºã§ã¯word2vecã§ããªãã¿ã®
ãçæ§ããŒãç·ãïŒã女ãïŒã女çã
ã**ã©ãŒã¡ã³ããŒãžã§ã³**ã§ãã£ãŠã¿ãããšæããŸãã
ãŽãŒã«ãšããŠã¯ã
ãã©ãŒã¡ã³ãïŒãè±éªšãïŒãã€ãµã€ã¿ã¯ãŒãïŒãäºéã
ã¿ããã«ãªãã°ãããªãšã
âãããªã€ã¡ãŒãžã§ãã
#ïŒ.ã¯ããã«
ä»åã¯ãããPythonãã©ãŒã¡ã³ã¬ãå¢ã«ãããããã·ãªãŒãºç¬¬ïŒåŒŸã§ãã
第ïŒåŒŸã§ã¹ã¯ã¬ã€ãã³ã°ããããŒã¿ãçšããŠã¢ãã«æ§ç¯ããŠãããŸãã
第ïŒåŒŸïŒãPythonãã©ãŒã¡ã³ã¬ãå¢ã«ããã¬ãå¢ã®ããã®é£ã¹ãã°ã¹ã¯ã¬ã€ãã³ã°
第ïŒåŒŸïŒ æ¬èšäº
第ïŒåŒŸïŒ[ãPythonãðæ©æ¢°åŠç¿ã§ãé ããååºããæ¢ããŠã¿ããïŒãããŠå®éã«è¡ã£ãŠã¿ãïŒð]
(https://qiita.com/toshiyuki_tsutsui/items/b3ac8fd1b300c3404508)
#ïŒ.Word2vecãšã¯ïŒ
ãããã**word2vecã䜿ããšäœãè¯ãã®ãïŒ**ã«ã€ããŠç°¡åã«è§£èª¬ããŸãã
word2vecã¯ã倧éã®ããã¹ãããŒã¿ãè§£æããŠãååèªã®æå³ããã¯ãã«è¡šçŸåããææ³ã§ããåèªããã¯ãã«åããããšã§ã
åèªéã®æå³ã®è¿ããèšç®
åèªéã®æå³ãè¶³ãããåŒããã
ãšããããšãå¯èœã«ãªããŸãã
ã£ãŠãã©ãããããšïŒ
ç§ã®ãããªååŠè ã®ããã«å°ã説æããŸããšãäŸãã°ä»£è¡šçãªã©ãŒã¡ã³ãã§ãŒã³åºãäžé¢šå ããã俺æµå¡©ã©ãŒã¡ã³ãããã¯ãã«ã§è¡šçŸãããš
äžé¢šå ïŒ(0.4,0.1,0.9,0.4)
俺æµå¡©ã©ãŒã¡ã³ïŒ(0.5,0.2,0.3,0.4)
ãšãªã£ããšããŸãã
ãã®äºã€ã®ãã¯ãã«ã®ã³ãµã€ã³é¡äŒŒåºŠãèšç®ããŸãã
ã³ãµã€ã³é¡äŒŒåºŠã¯äžè§é¢æ°ã®æ®éã®ã³ãµã€ã³ã®éãã1ã«è¿ããã°é¡äŒŒããŠããã0ã«è¿ããã°é¡äŒŒããŠããªããšè§£éã§ããŸãã
åèïŒhttp://www.cse.kyoto-su.ac.jp/~g0846020/keywords/cosinSimilarity.html
ãäžé¢šå ããã¯ãã«ãšã俺æµå¡©ã©ãŒã¡ã³ããã¯ãã«ã®ã³ãµã€ã³é¡äŒŒåºŠãèšç®ãããš0.83ãšãªããé¡äŒŒåºŠãé«ããšãããŸãããã®æè¡ãå¿çšãããšã©ãŒã¡ã³å±Aã«è¿ãã©ãŒã¡ã³å±Bãã¬ã³ã¡ã³ãããããšãå¯èœãšãªããŸãã
次ã«ãåèªéã®æå³ãè¶³ãããåŒãããããããšãã§ããŸãã
ä»åºŠã¯ãè±éªšããå¡©ããã¯ãã«ãçšæããŠã¿ãŸããã
äžé¢šå ïŒ(0.4,0.1,0.9,0.4)
俺æµå¡©ã©ãŒã¡ã³ïŒ(0.5,0.2,0.3,0.4)
è±éªšïŒ(0.1,0.0,0.8,0.2)
å¡©ïŒ(0.2,0.1,0.2,0.3)
å
ã»ã©ãäžé¢šå ã ãš ã俺æµå¡©ã©ãŒã¡ã³ãã®é¡äŒŒåºŠãé«ãã£ãããšããã
ãäžé¢šå ã ãã¯ãã«ããããè±éªšããã¯ãã«æåãæããŠãå€ããã«ãå¡©ããã¯ãã«æåãå ãããšã
ãäžé¢šå ã - ãè±éªšã + ãå¡©ã â ã俺æµå¡©ã©ãŒã¡ã³ã
ãšãã£ãèšç®ãã§ããŸãã
詳现ãªã¢ã«ãŽãªãºã ã¯ãäžèšãåèã«ãªãããšæããŸãã
https://deepage.net/bigdata/machine_learning/2016/09/02/word2vec_power_of_word_vector.html
#ïŒ.ã¢ãã«æ§ç¯ãŸã§ã®æµã
Word2vecã§ã¢ãã«æ§ç¯ããæé ããã£ãããå
šäœåãããããšçè§£ãæ©ãã§ãã
ååŠè
ã®æ¹ã¯ãäžèšã®èšäºã§èªç¶èšèªåŠççšèªãäºç¿ããŠããã ãããšã¹ã ãŒãºã§ãã
åèïŒhttps://qiita.com/yura/items/6c1481ca652d3d131e47
##â MeCabã®ã€ã³ã¹ããŒã«
MeCabãšã¯æ¥æ¬èªã®åœ¢æ
çŽ è§£æåšã®ããšã§ãã
åèïŒhttps://qiita.com/menon/items/f041b7c46543f38f78f7
##â¡Mecabã®èŸæžãNEologdã«å€æŽ
NEologdãšã¯MeCabçšã®ã·ã¹ãã èŸæžã®ããšãæããWebäžããåŸãæ°èªã«å¯Ÿå¿ããŠããŸãã
ååŠè
ã®èºããããç¹ã¯ãMeCabã€ã³ã¹ããŒã«åŸã«èŸæžãããã©ã«ãããNEologdã«å€æŽãããšããã§ãã誰ãããä¹ãè¶ããå£ã§ãã®ã§ãå
人ã®qiitaèšäºãåèã«ãããªã©ããŠã¯ãªã¢ããŠãã ããïŒïŒç§ã¯åæ¥ãããè²»ãããŸããç¬ïŒ
åèïŒhttps://qiita.com/menon/items/f041b7c46543f38f78f7
â»2019å¹ŽïŒææç¹ã§ã¯ãmecab-python3 0.996.1ã§ã¯parseToNodeãåèªã§ã¯ãªãæç« å
šäœãåºãäžå
·åãå ±åãããŠããŸããmecab-python3ã®ããŒãžã§ã³ãäžããŠå¯Ÿå¿ããå¿
èŠãããã¿ããã§ãã
pip uninstall mecab-python3
pip install mecab-python3==0.7
https://teratail.com/questions/164787
##â¢ã³ãŒãã¹ãäœæ
ã³ãŒãã¹ãšã¯ãèªç¶èšèªã®æç« ãæ§é åããŠãŸãšãããã®ãæããŸãã
å£ã³ãããŒã¿ãMecabã§åãã¡æžãããŠã³ãŒãã¹ãäœæããŸãã
ããããã¢ããªã³ã°ããããã®åŠç¿ããŒã¿ãšãªããŸãã
ä»å䜿çšããå£ã³ãããŒã¿ã¯ååã¹ã¯ã¬ã€ãã³ã°ã§ååŸãããã®ã§ãã
##â£ã¢ãã«æ§ç¯
gensimã䜿ã£ãŠWord2Vecãå®è£
ããŸãã
ã³ãŒãã¹ãWord2vecã«åãŸããã ãã§ãã
詳ããã¯ãœãŒã¹ã³ãŒããã芧ãã ããã
#ïŒ.ãœãŒã¹ã³ãŒã
import numpy as np
import pandas as pd
import pickle
from gensim.models import word2vec
import MeCab
tagger = MeCab.Tagger('-Owakati -d /usr/local/lib/mecab/dic/mecab-ipadic-neologd')#ã¿ã°ã¯MeCab.TaggerïŒneologdèŸæžïŒã䜿çš
tagger.parse('')
def tokenize_ja(text, lower):
node = tagger.parseToNode(str(text))
while node:
if lower and node.feature.split(',')[0] in ["åè©","圢容è©"]:#åãã¡æžãã§ååŸããåè©ãæå®
yield node.surface.lower()
node = node.next
def tokenize(content, token_min_len, token_max_len, lower):
return [
str(token) for token in tokenize_ja(content, lower)
if token_min_len <= len(token) <= token_max_len and not token.startswith('_')
]
#åŠç¿ããŒã¿ã®èªã¿èŸŒã¿
df = pd.read_csv('../output/tokyo_ramen_review.csv')
df_ramen = df.groupby(['store_name','score','review_cnt'])['review'].apply(list).apply(' '.join).reset_index().sort_values('score', ascending=False)
#ã³ãŒãã¹äœæ
wakati_ramen_text = []
for i in df_ramen['review']:
txt = tokenize(i, 2, 10000, True)
wakati_ramen_text.append(txt)
np.savetxt("../work/ramen_corpus.txt", wakati_ramen_text, fmt = '%s', delimiter = ',')
# ã¢ãã«äœæ
word2vec_ramen_model = word2vec.Word2Vec(wakati_ramen_text, sg = 1, size = 100, window = 5, min_count = 5, iter = 100, workers = 3)
#sgïŒ0: CBOW, 1: skip-gramïŒ,sizeïŒãã¯ãã«ã®æ¬¡å
æ°ïŒ,windowïŒåŠç¿ã«äœ¿ãååŸã®åèªæ°ïŒ,min_countïŒnåæªæºç»å Žããåèªãç Žæ£ïŒ,iterïŒãã¬ãŒãã³ã°ååŸ©åæ°ïŒ
# ã¢ãã«ã®ã»ãŒã
word2vec_ramen_model.save("../model/word2vec_ramen_model.model")
ããã§ã¢ãã«æ§ç¯ã¯å®äºã§ãïŒ
#ïŒ.ââã«æãè¿ãåèªã¯â³â³
ããããããããããæ¬çªã§ãïŒ
ã§ã¯ãæ©é most_similar()ã§ãããåèªã«æãè¿ããåèªã調ã¹ãŠã¿ãŸãã
**é£ã¹ãã°å£ã³ãã³ãŒãã¹modelã® "åšå" ã宿ããã ãããã«ãäžè¬çãªwikipediaã³ãŒãã¹**ã§ã¢ãã«æ§ç¯ãããã®ãšæ¯èŒããŠãããŸãã
âwikipediaã³ãŒãã¹ã®äœãæ¹â
https://qiita.com/kenta1984/items/93b64768494f971edf86
ã§ã¯ãæåã«å§çåŒçãªæãã§æå
¥ããã¯ãŒãã¯ããã¡ãïŒ
$\huge{ã山岞ã}$
ã€ã麺ã®çã¿ã®èŠªã§ãããã€ã麺çã®ã¬ãžã§ã³ãofã¬ãžã§ã³ãã§ããæ±æ± è¢å€§åè»å åºäž»ã山岞ããããæ¬æã蟌ããŠãæåãããŠããã ããŸããã
ã§ã¯ããŸãã¯**wikipediaã³ãŒãã¹model**ã«æå
¥ããŸãã
wikipediaã«ãããã山岞ãã«æãè¿ããåèªã衚瀺ãããŸãã
# ã¢ãã«ã®ããŒã
wikipedia_model = word2vec.Word2Vec.load("../model/full_wikipedia.model")
wikipedia_model.most_similar("山岞")
>>>
[('最å²', 0.4181003272533417),
('ç¯å®', 0.4008517265319824),
('å·¥è€', 0.39781779050827026),
('å
è»', 0.3838202953338623),
('æéš', 0.3719743490219116),
('岡å
', 0.3716839551925659),
('åæ²¢', 0.36764824390411377),
('å è€', 0.36382099986076355),
('æè€', 0.362610787153244),
('æšå³¶', 0.36078932881355286)]
æãããŠãããã¯ã¬ãžã§ã³ãã山岞ãããã«è¿ããã¯ãŒããšãããã§ããããããã¡ããåŠã§ãã
äžè¬çãª**wikipediaã³ãŒãã¹model**ã§ã¯ãã¬ãžã§ã³ãã山岞ããããšé¢é£æ§ã®ãªãåèªã䞊ã³ãŸããã
ç¶ããŠãç§ã䞹粟ã蟌ããŠäœã£ã**é£ã¹ãã°å£ã³ãã³ãŒãã¹model**ã«æå ¥ããŠã¿ãŸãã»ã»ã»
# ã¢ãã«ã®ããŒã
word2vec_ramen_model =word2vec.Word2Vec.load("../model/word2vec_ramen_model.model")
word2vec_ramen_model.most_similar("山岞")
>>>
[('æ±æ± è¢å€§åè»', 0.6612457036972046),
('æ±æ± è¢', 0.6129428744316101),
('山岞äžé', 0.5511749982833862),
('倧åè»', 0.5429054498672485),
('æ»éå·', 0.5334186553955078),
('ãããã°', 0.5152110457420349),
('äžžä¿¡', 0.5020366907119751),
('ãã¹ã¿ãŒ', 0.48960864543914795),
('代ã
æšäžå', 0.48622703552246094),
('1955幎', 0.4847238063812256)]
"$\huge{ãã¿ã³ã¬ïŒ}$"
ããã¯æå¥ãªãã§ã¬ãžã§ã³ãã山岞ãããã«é¢é£ãã倧åè»ã®ã¯ãŒãã䞊ã³ãŸããïŒãããã§ããã
æ±æ± è¢å€§åè»ïŒhttps://tabelog.com/tokyo/A1305/A130501/13045828/
次ã«**ãäºéãã«ã€ããŠãwikipediaã³ãŒãã¹modelãšé£ã¹ãã°å£ã³ãã³ãŒãã¹model**ã§çµæãæ¯èŒããŠã¿ãŸãã
# ã¢ãã«ã®ããŒã
wikipedia_model = word2vec.Word2Vec.load("../model/full_wikipedia.model")
wikipedia_model.most_similar("äºé")
>>>
[('äžé', 0.4573158621788025),
('次é', 0.44129377603530884),
('äžé', 0.43840569257736206),
('æŽæ³', 0.3729743957519531),
('åé', 0.33052968978881836),
('倪ç°å', 0.33035555481910706),
('倪é', 0.32737094163894653),
('åé ', 0.3245748281478882),
('æ±ç¥', 0.3236039876937866),
('ä¿¡äžé', 0.31906336545944214)]
決ããŠééããšããããã§ã¯ãªãã®ã§ããã»ã»ã»
ç§ãæ±ããŠãã**ãäºéã**ã¯ã
# ã¢ãã«ã®ããŒã
word2vec_ramen_model=word2vec.Word2Vec.load("../model/word2vec_ramen_model.model")
word2vec_ramen_model.most_similar("äºé")
>>>
[('ã©ãŒã¡ã³äºé', 0.7518627643585205),
('äºéç³»', 0.7041865587234497),
('ã€ã³ã¹ãã€ã¢', 0.6942269802093506),
('äžéæ¯', 0.6394986510276794),
('ã¡ã°ãž', 0.6040332317352295),
('ã€ãµã€', 0.5899537205696106),
('ä¹³å', 0.5867205858230591),
('çŽç³»', 0.5784134268760681),
('è±äº', 0.5678684711456299),
('äžä¹æ±', 0.567740261554718)]
"$\huge{ãŸãã«ããã§ããïŒ}$"
解説ãããšã
**ãäžéæ¯ãã¯ã©ãŒã¡ã³äºéäžéæ¯åºãæãããã¡ã°ãã**ã¯ã©ãŒã¡ã³äºéç®é»åºã®ããšã§ãã
**ãè±äºã**ãäºéç³»ã®ãåºã§ããã
ã©ãŒã¡ã³è±äºïŒhttps://tabelog.com/tokyo/A1326/A132602/13164704/
ãã®ããã«æ¯èŒãããšé£ã¹ãã°å£ã³ãã§äœã£ãã³ãŒãã¹ã®æèœããéç«ã€çµæãšãªããŸããïŒ
#ïŒ.è¶³ãç®ãåŒãç®ã«ãã£ã¬ã³ãž
ããããã¯ã¯ã€ãºåœ¢åŒã§é²ããŠãããŸãã
é¡äŒŒåºŠãäžçªé«ãã¯ãŒããäºæ³ããŠã¿ãŠãã ããã
çããããèŠããŠããŸãã¯ããæå¬ãšããããšã§(^^;
##â åçŽç·š
ãã©ãŒã¡ã³ã ïŒ ãåæµ·éã = ãïŒïŒïŒã
åæµ·éã®ãåœå°ã©ãŒã¡ã³ãšãã£ãããã¡ããïŒ
word2vec_ramen_model.most_similar(positive=[u"ã©ãŒã¡ã³",u"åæµ·é"])
>>>
[('å³åã©ãŒã¡ã³', 0.5989491939544678),
('ããŒãã', 0.5617096424102783),
('æå¹', 0.5409911870956421),
('ãã', 0.5338827967643738),
('ã©ãŒã¡ã³å±', 0.5107439756393433),
('逿²¹ã©ãŒã¡ã³', 0.5045751929283142),
('ãã¡ã', 0.5027260780334473),
('æ±äº¬', 0.49508172273635864),
('å
šæ¥ç©º', 0.4923136830329895),
('ã«ãªã£', 0.48750197887420654)]
æ£è§£ã¯å³åã©ãŒã¡ã³ã§ãã
##â¡äžçŽç·š
ãã©ãŒã¡ã³ã ïŒ ã忥µã = ãïŒïŒïŒã
**æ¿èŸã©ãŒã¡ã³**ã®å®çªãšããã°ãããïŒ
çå¬ã«é£ã¹ãŠãè¶
æ±ã ãã«ãªã匷çãªã©ãŒã¡ã³ã§ãã
word2vec_ramen_model.most_similar(positive=[u"ã©ãŒã¡ã³",u"忥µ"])
>>>
[('äžæ¬', 0.5654767751693726),
('ãã', 0.5430104732513428),
('17å', 0.5270459651947021),
('æ¿èŸ', 0.5181963443756104),
('walker', 0.5173792243003845),
('èŸã', 0.5158922672271729),
('èŸã', 0.5034847259521484),
('ææŠ', 0.49608075618743896),
('æš©åš', 0.48732683062553406),
('420g', 0.4867885708808899)]
ããã§ããæ£è§£ã¯**ãäžæ¬ã**ã§ããïŒ
**ã忥µããšããã°ãäžæ¬ã**ãšåå°çã«çããããããªãã¯ã¬ãå¢ã®çŽ è³ªããã§ãã
èå€ã¿ã³ã¡ã³äžæ¬ïŒhttps://tabelog.com/tokyo/A1322/A132203/13004380/

ãã©ãŒã¡ã³ã ïŒ ã忥µã ãŒ ãæ¿èŸã = ãïŒïŒïŒã
word2vec_ramen_model.most_similar(positive=[u"ã©ãŒã¡ã³",u"忥µ"], negative=[u"æ¿èŸ"])
>>>
[('ã©ãŒã¡ã³å±', 0.5003402233123779),
('ã©ãŒã¡ã³åº', 0.492368221282959),
('17å', 0.4728690981864929),
('award', 0.4705893397331238),
('人æ°', 0.47028398513793945),
('æ¿æŠ', 0.463814914226532),
('ãžã§ãŒ', 0.441331684589386),
('ã«ãªã£', 0.44119763374328613),
('æç«', 0.4396098256111145),
('å°éåº', 0.4389714002609253)]
ãªããšãæ®éã®ã©ãŒã¡ã³å±ã«æãäžãã£ãŠããŸããŸãã(ÂŽã»Ïã»`)
##â¢äžçŽç·š
ãå
åèã - ãæ±äº¬ã©ãŒã¡ã³ã¹ããªãŒãã ïŒ ãæŸæžã = ãïŒïŒïŒã
ç§ã¯ãïŒæé䞊ã³ãŸããã
word2vec_ramen_model.most_similar(positive=[u"å
åè",u"æŸæž"], negative=[u"æ±äº¬ã©ãŒã¡ã³ã¹ããªãŒã"])
>>>
[('å¯ç°', 0.46717607975006104),
('å¹³äº', 0.4340380132198334),
('éžææš©', 0.38690489530563354),
('å§åç', 0.3818015456199646),
('å
¥ãæ¿ã', 0.3796331286430359),
('éå¹', 0.3761281669139862),
('ãã', 0.3747977018356323),
('ãããã¥ãŒã¹', 0.3694424629211426),
('ããã', 0.3633619248867035),
('æåŒ·', 0.3624943792819977)]
æ£è§£ã¯**ãå¯ç°ã**ã§ããïŒ
ãæ£åŒåç§°ã¯"ãšã¿ç°"ã§ããããšããããã³ããå
¥ãããã§ããã
解説ïŒã€ã麺çã®ã«ãªã¹ãçååšãå
åèãïŒ æ±äº¬ã©ãŒã¡ã³ã¹ããªãŒãããããæ±äº¬ã©ãŒã¡ã³ã¹ããªãŒãããã¯ãã«ãåŒããŠãæŸæžããã¯ãã«ãå ãããšãæŸæžã«ãããã€ã麺çã®ã«ãªã¹ãçååšããšã¿ç°ããšãªããŸãã
äžè¯è麊 ãšã¿ç°ïŒhttps://tabelog.com/chiba/A1203/A120302/12000422/
âã®çµæãã¡ãã£ãšåŸ®åŠãªã®ã¯ãä»åã®ã³ãŒãã¹ã¯æ±äº¬ã©ãŒã¡ã³ã®å£ã³ãã®ã¿ã§äœæããŠããããããšã¿ç°ïŒ æŸæžã«ã€ããŠã®ããŒã¿ãè¶³ããªãã£ãã®ã ãšæãããŸãã
#ïŒ.ãŸãšã
ä»åã¯ååºã©ãŒã¡ã³å±ã®å£ã³ãããŒã¿ã䜿çšããŠword2vecã§éãã§ã¿ãŸããã
ååºã®å£ã³ããªãã§ã¯ã®"ãšããž"ã®å¹ããã¢ãã«ãã§ããŠè峿·±ãçµæãåŸãããšãã§ããŸããããã¢ãã«ã®è¯ãæªãã¯ããŒããã¬ãå¢ã«ããå€å®ã§ããªãã®ãæ©ã¿ã©ããã§ãã
èšäºã®åé ã§ã
ãã©ãŒã¡ã³ãïŒãè±éªšãïŒãã€ãµã€ã¿ã¯ãŒãïŒãäºéã
ã«ãªãã®ã§ã¯ïŒãšæšæž¬ããŸããããå
šç¶ããŸããããŸããã§ãããåå ã¯ã**ãäºéãã¯ãã©ãŒã¡ã³ãã«ããè±éªšã**ã«ãé¡äŒŒããŠããªãã£ãããã§ãã
ãäºéã¯ã©ãŒã¡ã³ã§ã¯ãªããäºéãšããé£ã¹ç©ã§ãããã
ãšããåèšã¯ãæ¬åœã«ãããªã®ãããããŸãããã
以äžã
次åã¯ãä»åäœæããã©ãŒã¡ã³ã¢ãã«ãå¿çšããŠãé ããååºãã¬ã³ã¡ã³ãã§çºæã«ãã£ã¬ã³ãžããŸãã
#9.仲éãæ¢ããŠããŸã
äœè«ã§ãããç§ã®é£ã®åžã§äžç·ã«æ©æ¢°åŠç¿ã¿ã¹ã¯ãã¬ã·ã¬ã·åãçµããçŽ æµãªæ¹ãæ¢ããŠãããŸããæ°èã§ããã¹ããŒãããã¯ãšããããŒããŠã§ã¢ã®äžã§ããããŸã§ã«ãªã倿§ãªAIã»æ©æ¢°åŠç¿ã®å¯èœæ§ã暡玢ããŸãããïŒ
[AIã»æ©æ¢°åŠç¿ã»ãã€ãªã¡ããªã¯ã¹ãšã³ãžãã¢]
(https://open.talentio.com/1/c/bitkey/requisitions/detail/12992)
ãŸãBitkeyã§ã¯ãAIã»æ©æ¢°åŠç¿é åã«éããæ§ã
ãªããžã·ã§ã³ã§äžç·ã«åãã仲éãåéããŠããŸããããèå³ãããæ¹ã¯ãæ°è»œã«ãäžå ±ãã ããã
[åéããžã·ã§ã³äžèЧ]
(https://open.talentio.com/1/c/bitkey/requisitions/1372)