2021年度-基盤研究B-研究計画調書

1. ï׽ࣈ౳ð ݚڀ୅දऀ ॴଐݚڀػؔ ྩ࿨úïù÷ùøð೥౓ ج൫ݚڀʢ̗ʣʢҰൠʣݚڀܭըௐॻ ø÷ çù ùý ೥ ೔ ݄ ôçøçô ੩Ԭେֶ ෦ɹہ ৘ใֶ෦ ৬ ïϑϦΨφð ϠϚϞτçϢ΢εέ ߨࢣ ݚڀ՝୊໊ ൷൑తͳ΢Σϒ৘ใ୳ࡧΛ׆ੑԽͤ͞Δ৘ใΠϯλϥΫγϣϯ ࢯ໊ ೥౓ ઃඋඋ඼අ ফ໣඼අ ݚڀܦඅ çʢઍԁʣ ཱྀඅ ͦͷଞ ࢖༻಺༁ʢઍԁʣ ૯ܭ ݚçڀçܦçඅ ઍԁະຬͷ ୺਺͸੾Γ ࣺͯΔ ৽ن ྩ࿨ú೥౓ ྩ࿨û೥౓ ྩ࿨ü೥౓ ྩ࿨ý೥౓ ççççúóýĀ÷ ççççüóþù÷ ççççúóþù÷ ççççûóüù÷ çççççççç÷ ççççççÿÿ÷ ççççççøú÷ ççççøóø÷÷ ççççççĀý÷ ççççúóÿ÷÷ ççççùó÷÷÷ çççççççç÷ çççççççç÷ çççççççç÷ çççççççÿ÷ çççççççÿ÷ çççççççÿ÷ çççççççç÷ ççççøóý÷÷ ççççøóý÷÷ ççççùóø÷÷ çççççççç÷ ççççøó÷ÿ÷ ççççøó÷ÿ÷ ççççøó÷ÿ÷ çççççççç÷ ççççççýù÷ ççççççĀý÷ ççççççĀý÷ ççççøóùý÷ ։ࣔر๬ͷ༗ແ ৹ࠪ݁Ռͷ։ࣔΛر๬͢Δ çú൛ ݚڀܭը࠷ऴ೥౓લ೥౓Ԡื ôô ྩ࿨þ೥౓ çççøþóýü÷ ççççùóÿÿ÷ ççççççúþ÷ ççççýóû÷÷ ççççûóù÷÷ çççççççç÷ ਓ݅අ ँۚ ɾ ø ÷÷÷ø øúÿ÷ø ýù÷ù÷ ÷ü ੔ཧ൪߸ খ۠෼ ػؔ൪߸ ݚڀछ໨൪߸ Ԡื۠෼൪߸ খ۠෼ ΢Σϒ৘ใֶ͓ΑͼαʔϏε৘ใֶؔ࿈ ݚڀछ໨ ج൫ݚڀïĉð Ԡื۠෼ Ұൠ ྩ࿨ ࢁຊç༞ี
2. ࢁຊɹ༞ี ϠϚϞτɹϢ΢εέ ߨࢣ ৘ใֶ෦ ݚڀ૊৫ʢݚڀ୅දऀٴͼݚڀ෼୲ऀʣ ࢯ໊ʢ೥ྸʣ ॴଐݚڀػؔ ෦ہ ৬ ֶҐ ໾ׂ෼୲ ੩Ԭେֶ ത࢜ʢ৘ใֶʣ ݚڀ୅දऀ ݚڀܦඅ ΤϑΥ ïઍԁð ʔτ ïìð ôçùçô ݚ ڀ ୅ ද ऀ ߹ܭ ໊ ݚڀܦඅ߹ܭ ྩ࿨ú೥౓ ççççúóýĀ÷ ççççúóýĀ÷ çøü øúÿ÷øô÷üøôýù÷ù÷ô÷÷÷ø ççø ü÷ýùüûúø ʢúÿʣ
3. ༷ࣜ S ʵ̍̏ ݚڀܭըௐॻʢఴ෇ϑΝΠϧ߲໨ʣ ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̍ ᇶ┙◊✲㸦㹀㸧㸦୍⯡㸧㸯 㸯 ◊✲┠ⓗࠊ◊✲᪉ἲ࡞࡝ ᮏ◊✲ィ⏬ㄪ᭩ࡣࠕᑠ༊ศࠖࡢᑂᰝ༊ศ࡛ᑂᰝࡉࢀࡲࡍࠋグ㏙࡟ᙜࡓࡗ࡚ࡣࠊࠕ⛉Ꮫ◊✲㈝ຓᡂ஦ᴗ࡟࠾ࡅࡿᑂᰝཬࡧホ౯ ࡟㛵ࡍࡿつ⛬ࠖ㸦බເせ㡿㸯㸯㸯㡫ཧ↷㸧ࢆཧ⪃࡟ࡍࡿࡇ࡜ࠋ ᮏḍ࡟ࡣࠊᮏ◊✲ࡢ┠ⓗ࡜᪉ἲ࡞࡝࡟ࡘ࠸࡚ࠊ㸲㡫௨ෆ࡛グ㏙ࡍࡿࡇ࡜ࠋ ෑ㢌࡟ࡑࡢᴫせࢆ⡆₩࡟ࡲ࡜ࡵ࡚グ㏙ࡋࠊᮏᩥ࡟ࡣࠊ  ᮏ◊✲ࡢᏛ⾡ⓗ⫼ᬒࠊ◊✲ㄢ㢟ࡢ᰾ᚰࢆ࡞ࡍᏛ⾡ⓗࠕၥ࠸ࠖࠊ  ᮏ◊✲ࡢ┠ⓗ࠾ࡼࡧᏛ⾡ⓗ⊂⮬ᛶ࡜๰㐀ᛶࠊ  ᮏ◊✲࡛ఱࢆ࡝ࡢࡼ࠺࡟ࠊ࡝ࡇࡲ࡛᫂ࡽ࠿࡟ࡋࡼ࠺࡜ࡍࡿࡢ࠿ࠊ࡟ࡘ࠸࡚ලయ ⓗ࠿ࡘ᫂☜࡟グ㏙ࡍࡿࡇ࡜ࠋ ᮏ◊✲ࢆ◊✲ศᢸ⪅࡜࡜ࡶ࡟⾜࠺ሙྜࡣࠊ◊✲௦⾲⪅ࠊ◊✲ศᢸ⪅ࡢලయⓗ࡞ᙺ๭ࢆグ㏙ࡍࡿࡇ࡜ࠋ 㸦ᴫせ㸧  㸦ᮏᩥ㸧 ͤ␃ព஦㡯㸸 1. సᡂ࡟ᙜࡓࡗ࡚ࡣࠊ◊✲ィ⏬ㄪ᭩సᡂ࣭グධせ㡿ࢆᚲࡎ☜ㄆࡍࡿࡇ࡜ࠋ 2. ᮏᩥ඲యࡣ㸯㸯࣏࢖ࣥࢺ௨ୖࡢ኱ࡁࡉࡢᩥᏐ➼ࢆ౑⏝ࡍࡿࡇ࡜ࠋ 3. ྛ㡫ࡢୖ㒊ࡢࢱ࢖ࢺࣝ࡜ᣦ♧᭩ࡁࡣື࠿ࡉ࡞࠸ࡇ࡜ࠋ 4. ᣦ♧᭩ࡁ࡛ᐃࡵࡽࢀࡓ㡫ᩘࡣ㉸࠼࡞࠸ࡇ࡜ࠋ࡞࠾ࠊ✵ⓑࡢ㡫ࡀ⏕ࡌ࡚ࡶ๐㝖ࡋ࡞࠸ࡇ ࡜ࠋ 5. ᮏ␃ព஦㡯㸦ᩳయࡢᩥ❶㸧ࡣࠊ◊✲ィ⏬ㄪ᭩ࡢసᡂ᫬࡟ࡣ๐㝖ࡍࡿࡇ࡜ࠋ ᵝᘧ㹑㸫㸯㸱◊✲ィ⏬ㄪ᭩㸦ῧ௜ࣇ࢓࢖ࣝ㡯┠㸧 ʢ֓ཁʣ ຊݚڀ՝୊Ͱ͸ɼ൷൑తͳ΢Σϒ৘ใ୳ࡧΛଅਐɾ׆ੑԽͤ͞Δ৘ใΠϯλϥΫγϣϯٕज़ͷݚ ڀ։ൃΛߦ͏ɽ۩ମతʹ͸ҎԼͷݚڀ߲໨ʹऔΓ૊Ήɿ • ৘ใ୳ࡧߦಈΛ಺লɾվળ͢ΔͨΊͷ΢Σϒϒϥ΢β • φοδΛ༻͍ͨɼ৘ใਫ਼ࠪଶ౓Λܹࢗ͢Δݕࡧ݁ՌαϚϦͷੜ੒ • ൷൑త৘ใ୳ࡧΛϑΝγϦςʔγϣϯ͢Δʮ໰͍ʯͷର࿩తఏࣔ ৘ใͷ৴པੑΛҙ͍ࣝͯ͠ͳ͍Ϣʔβ΍ภͬͨ৘ใऔಘΛߦ͍ͬͯΔϢʔβ͕ɼۄੴࠞᔿͷ΢Σ ϒ͔Β൷൑తʹ৘ใΛऔಘ͢Δ͏͑ͰඞཁͱͳΔεΩϧͱଶ౓Λແཧͳࣗ͘વʹվળͰ͖ΔΑ͏ɼ ্ه 3 ߲໨ͷݚڀ։ൃʹऔΓ૊Ήɽ ʢຊจʣ ຊݚڀͷֶज़తഎܠ ΢Σϒ΍৘ใΞΫηεٕज़ͷൃలʹΑͬͯɼ୭΋͕ࣗ༝ʹ৘ใΛൃ৴ɾऔಘͰ͖ΔΑ͏ʹͳͬ ͨɽͦͷҰํͰɼ΢Σϒ৘ใͷ৴པੑ΍৘ใऔಘͷภΓ͕ࣾձ໰୊ʹͳ͍ͬͯΔɽίϩφՒʹ͓ ͍ͯ΋ɼ ʮ৽ܕίϩφ΢ΠϧεʹΑͬͯτΠϨοτϖʔύʔ͕ෆ଍͢Δʯͱ͍ͬͨϑΣΠΫχϡʔ ε͕ιʔγϟϧϝσΟΞ্Ͱ֦ࢄͨ͜͠ͱ͸ɼهԱʹ৽͍͠1ɽ ΢Σϒ৘ใͷ৴པੑ΍৘ใऔಘͷภΓʹ͔͔Δ໰୊͸ɼύʔιφϥΠθʔγϣϯٕज़ʹΑͬͯ ͞ΒʹਂࠁԽ͍ͯ͠Δɽྫ͑͹ɼ੓࣏τϐοΫʹରͯ͠ύʔιφϥΠθʔγϣϯΛ༻͍ͨ৘ใ഑ ৴͕ߦΘΕΔͱɼෆ͔֬ʹ΋͔͔ΘΒͣࣗ෼ͷڵຯ΍৴೦ʹҰக͢Δ৘ใΛݟΔػձ͕૬ରతʹ ૿͑Δ͜ͱʹΑΓɼݸਓͷภݟ͕ॿ௕͞Εɼࣾձͷ෼ۃԽ͕ਐΉ͜ͱ͕ࢦఠ͞Ε͍ͯΔɽ ͜ͷΑ͏ͳঢ়گʹରԠ͢ΔͨΊɼ৘ใݕࡧ΍σʔλ޻ֶͷ෼໺Ͱ͸ɼϑΣΠΫχϡʔεͷݕग़ ΞϧΰϦζϜΛؚΊɼ৘ใͷ඼࣭Λ༷ʑͳ֯౓͔ΒධՁɾ෼ੳ͢ΔΞϧΰϦζϜ͕ఏҊ͞Ε͖ͯ ͨɽ͜ΕΒܭࢉػʹΑΔ৘ใ඼࣭ͷࣗಈ෼ੳΞϓϩʔν͸ɼͦͷద༻ൣғ͕෼ੳʹඞཁͱͳΔ ࣄલ஌͕ࣝೖखͰ͖ΔυϝΠϯʹݶΒΕΔɽ·ͨɼΞϧΰϦζϜ͕ฦ͢ग़ྗ͸ɼ৘ใͷਖ਼͠͞Λ อূ͍ͯ͠ΔΘ͚Ͱ͸ͳ͍ɽ͞Βʹผͷ໰୊ͱͯ͠ɼܭࢉػ͕৘ใͷ඼࣭Λ൑அ͢ΔࡐྉΛఏ ڙͯ͠΋ɼͦΕΛద੾ʹར༻Ͱ͖ͳ͍/ར༻͠ͳ͍Ϣʔβ͕ଘࡏ͢Δɽਃ੥ऀ͸ɼ΢Σϒ৘ใͷ ৴པੑ൑அࢧԉγεςϜͷݚڀͰɼϢʔβ͸ධՁ஋ͷߴ͍จॻΛ໡໨తʹӾཡͨ͠ΓɼγεςϜ ͷఏڙ৘ใΛࣗ෼ͷ৴೦ʹҰக͢ΔΑ͏ʹ౎߹Α͘ղऍ͠ɼ৘ใͷۛຯΛߦΘͳ͍܏޲͕͋Δ͜ ͱΛ໌Β͔ʹ͍ͯ͠Δʢݚڀۀ੷ [10, 11]ʣ ɽ ଟ༷ͳ؍఺͔Βͷ৘ใऔಘΛࢧԉ͢Δݕࡧ݁Ռͷଟ༷Խٕज़͸ɼ৘ใऔಘͷภΓΛܰݮ͢Δ্ Ͱ༗ޮʹಇ͘Մೳੑ͕͋Δ2ɽ͔͠͠ɼ৘ใ୳ࡧϢʔβ͸ࣗ෼͕޷Ή৘ใΛ༏ઌతʹӾཡ͢Δʮબ ୒త஫ҙ܏޲ʯΛ༗͢ΔɽͦΕΏ͑ɼภͬͨ৘ใऔಘΛ͍ͯ͠ΔϢʔβͷࢹ໺Λ޿͛Δʹ͸ɼબ ୒త஫ҙ܏޲ͷଘࡏΛߟྀͨ͠৘ใΞΫηεγεςϜͷઃܭ͕ٻΊΒΕΔɽ 1૯຿লɿ ʮ৽ܕίϩφ΢Οϧεײછ঱ʹؔ͢Δ৘ใྲྀ௨ௐࠪʯ ɼ2020 ೥ 6 ݄ɽ 2Rafiei, K. Davood et al. Diversifying Web Search Results. In Proceedings of the 19th International Conference on World Wide Web (WWW 2010), pp.781–90, 2010.
4. ʲ̍ɹݚڀ໨తɼݚڀํ๏ͳͲʢ͖ͭͮʣ ʳ ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̎ ۄੴࠞᔿͷ΢Σϒ৘ใ͔Βੜ্͖͍ͯ͘ͰඞཁͱͳΔ৘ใΛऔಘ͢Δʹ͸ɼطଘͷݚڀϓϩδΣ ΫτͰߦΘΕ͍ͯΔΑ͏ͳɼܭࢉػ͕ਓؒͷ୅ΘΓʹ৘ใͷ඼࣭ධՁΛߦ͏Ξϓϩʔν͚ͩͰ͸ ෆे෼Ͱ͋Δɽถࠃϐϡʔݚڀॴ΋ใࠂ͍ͯ͠ΔΑ͏ʹɼ΢Σϒʹ͓͚Δ৘ใͷ৴པੑ/ภͬͨ ৘ใऔಘͷ໰୊Λղܾ͢Δʹ͸ɼ୯ͳΔٕज़తͳ໰୊Ͱ͸ͳ͘ɼ৘ใΛड͚औΔϢʔβͷ৘ใϦ ςϥγʔʹ΋য఺Λ౰ͯΔඞཁ͕͋ΔɽPost-truth ࣌୅ͱ͍͏ݴ༿ʹ৅௃͞ΕΔΑ͏ʹɼࣗ෼͕ ৴༻Ͱ͖Δ৘ใ͕ʮਖ਼͍͠৘ใʯͰ͋Δͱߟ͑Δ෩ை͕ڧ·͍ͬͯΔঢ়گʹ͓͍ͯ͸ɼ৘ใΞΫ ηεγεςϜΛ࢖͍͜ͳ͠ɼϢʔβࣗΒ৘ใΛ൷൑తʹ୳ࡧɾӾཡ͢ΔϦςϥγʔ͕ɼ͜Ε·Ͱ Ҏ্ʹڧ͘ٻΊΒΕ͍ͯΔɽܭࢉػʹΑΔ৘ใ඼࣭ͷධՁΞϓϩʔνͷෆ׬શੑΛิ͏ͨΊʹ΋ɼ Ϣʔβͷ൷൑తͳ৘ใ୳ࡧ׆ಈΛ׆ੑԽͤ͞ΔΞϓϩʔν͕ॏཁͱͳΔɽ ݚڀ՝୊ͷ֩৺Λͳֶ͢ज़తʮ໰͍ʯ ઌʹड़΂ֶͨज़తഎܠΛ౿·͑Δͱɼ΢ΣϒϢʔβͷ൷൑తͳ৘ใ୳ࡧ׆ಈΛ׆ੑԽͤ͞Δʹ ͸ɼҎԼͷΑ͏ͳֵ৽తͳֶज़తʮ໰͍ʯʹ౴͑Δඞཁ͕͋Δɽ͢ͳΘͪɼ ໰͍ 1 ΢Σϒ৘ใΛ൷൑తʹ୳ࡧɾӾཡ͠Α͏ͱ͍ͯ͠ͳ͍ɼ͋Δ͍͸ҙཉ͸͋ͬͯ΋Ͱ͖͍ͯ ͳ͍Ϣʔβͷ൷൑త৘ใ୳ࡧ׆ಈΛ׆ੑԽͤ͞Δʹ͸ɼͲͷΑ͏ͳ৘ใγεςϜɾΠϯλϥ ΫγϣϯΛઃܭ͢Δ͜ͱ͕Մೳ͔ʁ ਃ੥ऀ͕աڈʹߦͬͨݚڀͰ͸ɼ΢ΣϒݕࡧΤϯδϯͳͲͷ৘ใΞΫηεγεςϜΛ༻͍ͯ൷൑త ʹ΢Σϒ৘ใΛऩू͢Δʹ͸ɼ ʮ΢Σϒ৘ใͷ඼࣭Λ֬ೝ͢ΔͨΊͷํུʯ ʮ৘ใΞΫηεγεςϜΛ࢖ ͍͜ͳ͢εΩϧʯ ʮ൷൑తͰ͋Ζ͏ͱ͢Δଶ౓ʯ ͳͲɼ ൷൑తͳ৘ใ୳ࡧΛߦ͏ͨΊͷεΩϧతଆ໘ ͱଶ౓తଆ໘ͷ྆໘ͰͷڧԽ͕ඞཁͰ͋Δ͜ͱ͕໌Β͔ʹͳ͍ͬͯΔʢݚڀۀ੷ [8]ʣ ɽ͕ͨͬ͠ ͯɼຊݚڀ՝୊Ͱ͸্ه໰͍͔Β೿ੜͯ͠ɼҎԼͷ 2 ͭͷֶज़తʮ໰͍ʯΛઃఆ͢Δɽ ໰͍ 2 Ϣʔβ͕ࣗ਎ͷ΢Σϒ୳ࡧߦಈΛলΈͯɼ൷൑త৘ใ୳ࡧʹඞཁͳεΩϧΛແཧͳ͘ண࣮ ʹվળ͢Δʹ͸ɼͲͷΑ͏ͳ৘ใγεςϜɾΠϯλϥΫγϣϯΛઃܭ͢Ε͹Α͍͔ʁ ໰͍ 3 ΢Σϒ୳ࡧߦಈΛ๦͛Δ͜ͱͳ͘ɼϢʔβͷੵۃతͳ൷൑త৘ใ୳ࡧߦಈΛ༠ൃ͢Δʹ͸ɼ ͲͷΑ͏ͳ৘ใγεςϜɾΠϯλϥΫγϣϯΛઃܭ͢Ε͹Α͍͔ʁ ຊݚڀͷ໨త ຊݚڀ՝୊Ͱ͸ɼ৘ใΞΫηεγεςϜΛ༻͍ͯ৘ใΛ൷൑తʹਫ਼ࠪ͠ɼҙࢥܾఆʹ༗༻ͳ৘ใΛ ੵۃతʹऩू͢ΔߦಈΛ൷൑త৘ใ୳ࡧͱఆٛ͢Δɽຊ՝୊ͷ໨త͸ɼ৘ใͷ৴པੑΛҙ͍ࣝͯ͠ ͳ͍Ϣʔβ΍ภͬͨ ɾ ҆қͳ৘ใऔಘΛߦ͍ͬͯΔϢʔβ͕ɼ ۄੴࠞᔿͷ΢Σϒ৘ใ͔Βੜ͖͍ͯͨ͘ Ίʹඞཁͳ৘ใΛ൷൑తʹऔಘͰ͖ΔΑ͏ɼ ൷൑త৘ใ୳ࡧʹඞཁͱͳΔεΩϧ΍ଶ౓Λແཧͳ͘ ࣗવʹվળ͠ɼ൷൑త৘ใ୳ࡧΛଅਐ͢Δ৘ใΠϯλϥΫγϣϯٕज़Λ։ൃ͢Δ͜ͱͰ͋Δɽ ֶज़తಠࣗੑͱ૑଄ੑ ਤ 1: ຊݚڀ՝୊ͷҐஔ͚ͮ ৘ใݕࡧ΍σʔλ޻ֶͷ෼໺Ͱ͸ɼߴ ඼࣭ͳ৘ใΛޮ཰Α͘৘ใݕࡧɾӾཡ͢ Δ͜ͱΛࢧԉ͢΂͘ɼ(1) ৘ใͷ඼࣭ʹؔ ͢Δ༷ʑͳ෼ੳΞϧΰϦζϜͷݚڀʢਤ 1 ࠨԼʣ ɼ(2) ൑அࡐྉͷࣗಈू໿ɾՄ ࢹԽٕज़ʹؔ͢Δݚڀʢਤ 1 ӈԼʣ͕ ߦΘΕ͖ͯͨɽຊݚڀ՝୊ͷֶज़తಠ
5. ʲ̍ɹݚڀ໨తɼݚڀํ๏ͳͲʢ͖ͭͮʣ ʳ ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̏ ࣗੑ͸ɼ্ه (1) ͱ (2) ͷΞϓϩʔνʹΑͬͯ։ൃ͞Εͨ৘ใΞΫηεγεςϜΛ͏·͘ ࢖͍͜ͳ͢͜ͱ͕Ͱ͖ͳ͍ʮਓؒͷඇ߹ཧతଆ໘ʯΛܭࢉػΛ༻͍ͯվળ͠Α͏ͱ͢Δ఺ʹ͋Δɽ ͜ΕΛୡ੒͢ΔͨΊʹɼຊݚڀ՝୊Ͱ͸ɼਃ੥ऀͷઐ໳෼໺Ͱ͋Δ৘ใݕࡧʹՃ͑ͯɼೝ஌৺ཧ ֶɼߦಈܦࡁֶɼڭҭ޻ֶͷ஌ݟΛԣஅతʹ׆༻͢Δɽ͜ͷ఺ʹ͓͍ͯɼຊݚڀ՝୊͸૑଄ੑ͕ ͋Δͱߟ͑Δɽ ຊݚڀͰԿΛͲͷΑ͏ʹɼͲ͜·Ͱ໌Β͔ʹ͠Α͏ͱ͢Δͷ͔ ਤ 2: ຊݚڀ՝୊ͷϑϨʔϜϫʔΫɽ ਤ 2 ʹɼຊݚڀ՝୊ͷϑϨʔϜ ϫʔΫΛࣔ͢ɽຊݚڀ՝୊Ͱ͸ɼ൷ ൑త৘ใ୳ࡧʹඞཁͱͳΔεΩϧ΍ ଶ౓Λ޲্ͤ͞ɼϢʔβͷੵۃతͳ ൷൑త৘ใ୳ࡧΛଅਐ͢ΔͨΊʹɼ ʢ1ʣ಺লɼ ʢ2ʣφοδʢ҉໧తઆಘʣ ɼ ʢ3ʣϑΝγϦςʔγϣϯʢ໌ࣔతઆಘʣ ɼ ͷ؍఺͔Βݚڀ։ൃΛߦ͏ɽҎԼʹ۩ ମతͳݚڀ߲໨Λه͢ɽ ݚڀ߲໨ 1: ࣗ਎ͷ৘ใ୳ࡧߦಈΛ಺লɾվળ͢ΔͨΊͷ΢Σϒϒϥ΢β ਤ 3: ൷൑త৘ใ୳ࡧεΩϧͷఔ౓΍վળϙΠ ϯτΛ಺লͤ͞Δϒϥ΢βͷΠϝʔδɽ ൷൑త৘ใ୳ࡧͷεΩϧ΍ํུΛ಺লɾվળ ͢ΔͨΊͷϒϥ΢βʹؔ͢ΔݚڀΛߦ͏ɽ ҰൠʹɼεΩϧ΍ೳྗΛվળ͢ΔͨΊʹ͸ɼࣗ ෼ͷݱঢ়ΛৼΓฦΓվળ͢΂͖఺Λ೺Ѳ͢Δ͜ ͱ͕ॏཁͰ͋Δͱ͞Ε͍ͯΔɽͱ͜Ζ͕ɼ൷൑త ৘ใ୳ࡧʹ͍ͭͯ͸ɼࣗ෼ͷεΩϧͷఔ౓΍ߦ ಈ܏޲Λϝλೝ஌͢Δ͜ͱ͸༰қͰ͸ͳ͍ɽҰ ํͰɼϢʔβࣗ਎ͷγεςϜར༻ύλʔϯΛল ΈΔͨΊͷ৘ใ΍๬·͍͠ߦಈͷྫࣔʹΑͬͯɼ γεςϜͷར༻εΩϧ͕͢Δͱ͍͏ݚڀࣄྫ΋ ͍͔ͭ͘ใࠂ͞Ε͍ͯΔ34ɽ ਃ੥ऀ͸աڈͷݚڀͰɼ৘ใΞΫηεγες ϜΛ༻͍ͯ൷൑తʹ΢Σϒ৘ใΛ୳ࡧ͢ΔεΩ ϧΛධՁ͢ΔͨΊͷई౓ͱ࣭໰ࢴΛ։ൃͨ͠ʢݚ ڀۀ੷ [8]ʣ ɽ·ͨɼ΢ΣϒݕࡧΤϯδϯͷݕࡧ ϩάΛղੳ͢Δ͜ͱͰɼ൷൑తͳ৘ใ୳ࡧΛߦ ͏Ϣʔβʹಛ༗ͷ΢ΣϒݕࡧߦಈΛ໌Β͔ʹͭͭ͋͠Δʢݚڀۀ੷ [7]ʣɽ͜ΕΒͷ஌ݟΛ׆༻ ͯ͠ɼ൷൑త୳ࡧ͕Ͱ͖ΔϢʔβͷ۩ମతͳߦಈΛྨܕԽ্ͨ͠ͰɼϢʔβͷ΢Σϒ୳ࡧߦಈ Λ;·͑ͯɼ൷൑త৘ใ୳ࡧεΩϧͷఔ౓΍վળϙΠϯτΛՄࢹԽ͢Δϒϥ΢βΠϯλϑΣʔ εΛ։ൃ͢Δʢਤ 3ʣɽڭҭ޻ֶͷ෼໺Ͱ͸ɼϞνϕʔγϣϯΛอͪͳ͕ΒվળΛଓ͚ΔͨΊ ʹ͸ɼ ʮݱঢ়ͷϨϕϧΑΓ΋΍΍গ͠ߴ͍՝୊Λఏࣔ͢Δ͜ͱʯ ʮଞऀൺֱΑΓ΋աڈͷࣗ෼ ͱͷൺֱʹΑΔධՁʯ͕༗ޮͱ͞Ε͍ͯΔɽ͜ͷΑ͏ͳ஌ݟ΋౿·͑ͳ͕Βɼ΢Σϒ୳ࡧϢʔ β͕ϞνϕʔγϣϯΛอͪͳ͕Βɼࣗ਎ͷ൷൑త৘ใ୳ࡧߦಈΛ๬·͍͠ํ޲ʹվળ͢ΔͨΊͷ ಺লΠϯλϑΣʔεͷઃܭͱධՁΛߦ͏ɽ 3Harvey et al.: Learning by Example: Training Users with High-Quality Query Suggestions. In Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2015), pp. 133–42. 4Bateman et al.: The Search Dashboard: How Reflection and Comparison Impact Search Behavior. In Pro- ceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI 2012), pp. 1785–94.
6. ʲ̍ɹݚڀ໨తɼݚڀํ๏ͳͲʢ͖ͭͮʣ ʳ ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̐ ݚڀ߲໨ 2: φοδΛ༻͍ͨɼ৘ใਫ਼ࠪଶ౓Λܹࢗ͢Δݕࡧ݁ՌαϚϦͷੜ੒ʢ҉໧తઆಘʣ ਤ 4: ։ൃ͢ΔݕࡧαϚϦͷྫɽ ຊݚڀ߲໨Ͱ͸ɼ൷൑తͳ৘ใ୳ࡧ͕ඞཁͳγʔϯͷҰྫͱ ͯ͠ɼۄੴࠞᔿͷ΢Σϒҩྍ৘ใͷݕࡧʢྫɿࢢൢༀɾපӃݕ ࡧʣʹয఺Λ౰ͯΔɽͦͷ্Ͱɼ৘ใਫ਼ࠪଶ౓Λܹࢗ͠ɼ൷൑ తͳ৘ใ୳ࡧߦಈΛແཧͳࣗ͘વʹ༠ൃ͢ΔɼφοδΛར༻͠ ͨݕࡧ݁ՌαϚϦʢ֓ཁจʣͷੜ੒ʹؔ͢ΔݚڀΛߦ͏ɽ Ұൠʹɼ৘ใͷ৴པੑ΍࣭ΛධՁ͢Δ͜ͱʹڧ͘ಈػ͚ͮΒ Ε͍ͯͳ͍Ϣʔβ͸ɼ҉໧ͷ͏ͪʹʮݕࡧ݁ՌͷॱҐʯ΍ʮϨ Ϗϡʔͷ੕ͷ਺ʯͱ͍ͬͨද૚తͳख͕͔ΓͰ৘ใͷ࣭Λ൑அ ͯ͠͠·͏͜ͱ͕஌ΒΕ͍ͯΔɽͦͷͨΊɼ࣮͸க໋తͳܽ఺ ͕͋Δʹ΋͔͔ΘΒͣɼݕࡧ݁ՌॱҐ͕ߴ͍͚ͩͰ৘ใΛӏವ Έʹͯ͠͠·͏Մೳੑ͕͋Δɽ ຊݚڀ߲໨Ͱ͸ɼ൷൑త৘ใ୳ࡧ΁ͷಈػ෇͚ͷͨΊʹʮφο δʯͱʮଛࣦճආόΠΞεʯʹண໨͢Δɽφοδͱ͸ߦಈܦࡁֶͷ෼໺Ͱ஫໨ΛूΊ͍ͯΔ֓೦ Ͱɼબ୒ͷ༨஍Λ࢒͠ͳ͕Β΋ϢʔβΛಛఆͷબ୒ʹ༠ಋ͢ΔΞϓϩʔνͰ͋Δɽଛࣦճආό ΠΞε͸ɼϝϦοτΑΓ΋σϝϦοτΛେ͖͘ײ͡΍͍͢ͱ͍͏ਓؒͷੑ࣭Ͱ͋Δɽຊݚڀ߲໨ Ͱ͸ɼ·ͣҩྍ৘ใʢࢢൢༀɾපӃʣʹؔ͢ΔϨϏϡʔ৘ใʹରͯ͠ Attention ػߏΛ༻͍ͨਂ ૚ࣗવݴޠॲཧΛߦ͍ɼଛࣦճආόΠΞεΛײͤ͡͞Δ؍఺Λநग़͢Δํ๏Λ։ൃ͢Δɽͦͷ্ Ͱɼҩྍ৘ใͷݕࡧ࣌ʹɼଛࣦճආόΠΞεΛײͤ͡͞Δ؍఺Λ૊Έ͜Μͩݕࡧ݁ՌαϚϦͷઃ ܭɾධՁΛߦ͏ɽݕࡧ݁ՌॱҐ্͕ҐͰ͋ͬͨΓ૯߹ධՁΛࣔ͢੕ͷ਺͕ଟ͍පӃɾࢢൢༀͰ΋ɼ ಛఆͷ؍఺ͰଛࣦͷՄೳੑΛײͤ͡͞Δ͜ͱͰɼ৻ॏͳ৘ใਫ਼ࠪΛ༠ൃ͢Δ͜ͱΛ໨ࢦ͢ɽ ݚڀ߲໨ 3: ൷൑త৘ใ୳ࡧΛϑΝγϦςʔτ͢Δʮ໰͍ʯͷର࿩తఏࣔʢ໌ࣔతઆಘʣ ຊ߲໨Ͱ͸ɼ൷൑తʹ৘ใΛಡΈղ্͘Ͱɼௐ΂Δ΂͖৘ใΛਂ͘ௐ΂Δ͜ͱΛઆಘ͢ΔͨΊͷ ৘ใݕࡧΠϯλϥΫγϣϯʹ͍ͭͯݚڀΛߦ͏ɽ۩ମతʹ͸ɼ ʮΑ͘ߟ͑Δͱɼ͔֬ʹͦͷτϐοΫʹ ͍ͭͯ͸͋·Γਂ͘ߟ͍͑ͯͳ͔ͬͨͷͰɼ ௐ΂Δඞཁ͕͋Δʯ ͱࢥΘͤΔΑ͏ͳ໰͍͔͚จΛ΢Σϒ ݕࡧதʹର࿩తʹఏࣔ͠ɼ൷൑త৘ใ୳ࡧΛϑΝγϦςʔτ͢ΔϘοτΛ։ൃ͠ɼͦͷ༗ޮੑΛ ݕূ͢Δɽྫ͑͹ɼ ʮίϩφ΢Πϧε Ξϧίʔϧʯͱ͍͏ݕࡧτϐοΫʹରͯ͠͸ʮೱ౓ʯ͕ॏཁ ͳؔ࿈τϐοΫͱͯ͠ߟ͑ΒΕΔɽ͜ͷࡍɼΞϧίʔϧͳΒԿͰ΋Α͍ͱࢥ͍ͬͯΔϢʔβʹೱ ౓ͷҧ͍͕ফಟޮՌʹӨڹ͢Δ͜ͱΛؾ෇͔ͤΔͨΊʹɼϘοτ͕ʮফಟʹඞཁͳΞϧίʔϧ౓ ਺͸Ͳͷఔ౓ͩͱࢥ͏͔ʁʯͱ͍͏໰͍Λ౤͔͚͛Δ͜ͱΛ૝ఆ͍ͯ͠Δɽ ڭҭ޻ֶ΍ίʔνϯάͷ෼໺Ͱ͸ɼର৅ऀ͔ΒࣗൃతͳࢥߟΛҾ͖ग़͠ɼؾ͖ͮ΍ߦಈΛଅ͢ ʹ͸ʮ໰͍ͷσβΠϯʯ͕ॏཁͱ͞Ε͍ͯΔ5ɽຊݚڀ߲໨Ͱ͸ɼಛఆͷτϐοΫʹ͍ͭͯݕࡧ͠ ͍ͯΔ΢ΣϒϢʔβͷϖʔδӾཡཤྺ͔ΒɼϢʔβ͕Ӿཡͨ͠τϐοΫͷ৘ใྔΛܭଌ͠ɼ໢ཏ ͍ͯ͠ͳ͍αϒτϐοΫΛʮਂ͘ௐ΂ΒΕ͍ͯͳ͍τϐοΫʯͱͯ͠நग़͢Δɽͦͷ্Ͱɼਂ͘ ௐ΂ΒΕ͍ͯͳ͍τϐοΫͷݕࡧΛଅ͢ʮྑ͍໰͍ʯΛ΢Σϒ͔Βൃݟ͠ɼ΢ΣϒݕࡧதͷϢʔ βʹϘοτΤʔδΣϯτΛ༻͍ͯର࿩తʹʮ໰͍ʯఏࣔ͢Δɽ Ұൠʹʮྑ͍໰͍ʯͰ͋Ε͹ɼճ౴ऀ͸ࢥߟ͕ଅਐ͞ΕɼࢥΘͣ౴͑ͨ͘ͳΔ΋ͷͰ͋Δɽਃ ੥ऀ͕༧උతʹߦͬͨௐࠪͰ͸ɼஶ໊ͳ࣭໰Ԡ౴αΠτͷ 1 ͭͰ͋Δ Yahoo!஌ܙାʹ౤ߘ͞Εͨ ࣭໰ʹ͸ɼଟ͘ͷϢʔβ͕ճ౴ͨ͘͠ͳΔΑ͏ͳྑ͍࣭໰͕͋Γɼྑ͍࣭໰͕ߦΘΕͨ৔߹ɼ࣭ ໰ऀͱճ౴ऀͷؒͰؾ͖͕ͮ͋ͬͨ͜ͱΛࣔࠦ͢Δٞ࿦͕ߦΘΕΔ܏޲ʹ͋Δ͜ͱ͕͏͔͕͑ͨ ʢݚڀۀ੷ [3]ʣ ɽͦ͜Ͱɼຊݚڀ՝୊Ͱ͸ɼ࣭໰ʹू·ͬͨճ౴਺΍࣭໰ͷݴޠతಛ௃ɼ࣭໰ऀͱ ճ౴ऀͷΠϯλϥΫγϣϯΛ෼ੳ͢Δ͜ͱͰɼ࣭໰Ԡ౴αΠτ͔ΒݕࡧτϐοΫʹؔ࿈͢Δʮྑ ͍࣭໰ʢ໰͍ʣ ʯΛൃݟ͢Δٕज़Λ։ൃ͢Δɽ·ͨɼऩूͨ͠ʮྑ͍໰͍ʯʹਂ૚ֶशΛద༻͠ɼ ೚ҙͷτϐοΫʹରͯ͠ʮྑ͍໰͍ʯΛࣗಈੜ੒͢ΔΞϧΰϦζϜͷݕ౼΋ߦ͏ɽ 5҆ࡈ༐थɾԘ੉ོ೭ஶɿ ʮ໰͍ͷσβΠϯʯ ɼֶܳग़൛ࣾʢ2020ʣ ɽ
7. ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̑ ᇶ┙◊✲㸦㹀㸧㸦୍⯡㸧㸳 㸰 ᮏ◊✲ࡢ╔᝿࡟⮳ࡗࡓ⤒⦋࡞࡝ ᮏḍ࡟ࡣࠊ  ᮏ◊✲ࡢ╔᝿࡟⮳ࡗࡓ⤒⦋࡜‽ഛ≧ἣࠊ  㛵㐃ࡍࡿᅜෆእࡢ◊✲ືྥ࡜ᮏ◊✲ࡢ఩⨨࡙ࡅࠊ࡟ࡘ࠸࡚㸯㡫௨ ෆ࡛グ㏙ࡍࡿࡇ࡜ࠋ ຊݚڀͷண૝ʹࢸͬͨܦҢͱ४උঢ়گ ਃ੥ऀ͸ɼ΢Σϒ৘ใͷ৴པੑ൑அࢧԉʹؔ͢ΔݚڀΛߦ͖ͬͯͨɽ͜ͷաఔͰɼ৘ใਫ਼ࠪʹ ಈػ͚ͮΒΕ͍ͯͳ͍৔߹͸ɼ৘ใͷ৴པੑΛ൑அ͢ΔͨΊʹ༗ӹͳ৘ใΛఏڙͯ͠΋ɼϢʔβ ͸ͦΕΒͷ৘ใΛ͏·͘׆༻Ͱ͖ͳ͍/͠ͳ͍ͱ͍͏໰୊͕໌Β͔ʹͳͬͨʢۀ੷ [11, 10]ʣ ɽҰ ํͰɼߴ඼࣭ͳ৘ใΛऔಘͰ͖Δ΢ΣϒݕࡧϢʔβ͸ɼ൷൑తࢥߟΛ΋ͬͯ৘ใΛ٬؍తʹಡΈ ղ͜͏ͱ͢Δଶ౓ͱํུΛ༗͍ͯ͠Δ͜ͱ͕໌Β͔ʹͳͬͨʢۀ੷ [7]ʣ ɽ͜ΕΒ஌ݟ͔Βɼੜ͖ ͍ͯͨ͘Ίʹඞཁͳ΢Σϒ৘ใΛϢʔβ͕औࣺબ୒Ͱ͖ΔΑ͏ʹͳΔʹ͸ɼ ʮ൷൑త৘ใ୳ࡧʹඞ ཁͱͳΔεΩϧ΍ଶ౓Λ޲্ͤ͞ΔΞϓϩʔν͕ඞཁͰ͋Δʯͱ͍͏໰୊ҙࣝʹࢸͬͨɽ ୯ಠͰ໧ʑͱߦ͏͜ͱ͕ଟ͍΢Σϒ৘ใ୳ࡧʹ͓͍ͯ͸ɼ ࣗ਎ͷ৘ใ୳ࡧߦಈͷળ͠ѱ͠΍ռΊΔ ΂͖఺Λࢦఠ͞ΕΔػձʹ๡͍͠ɽͦͷͨΊɼ൷൑తʹ৘ใΛಡΈղͨ͘ΊʹඞཁͱͳΔεΩϧ ΍ଶ౓Λվળ͢Δ͜ͱ͸೉͍͠ɽਃ੥ऀ͸ɼ൷൑త৘ใ୳ࡧʹඞཁͱͳΔεΩϧΛධՁ͢ΔͨΊ ͷࢦඪͷ։ൃΛߦ͖ͬͯͨʢۀ੷ [8]ʣ ɽݱࡏɼͦΕΒࢦඪΛ༻͍ͯɼ࣮ࡍͷ৘ใݕࡧɾӾཡߦಈ ͔ΒεΩϧͷଟՉΛධՁ͢Δख๏Λݚڀ։ൃதͰ͋Δɽຊݚڀ՝୊ʹ͓͍ͯ͸ɼͦΕΒ։ൃதͷ ख๏Λ༻͍ͯɼ ʮεΩϧͷվળ఺Λ಺লͤ͞ɼվળͷͨΊͷ۩ମతͳৼΔ෣͍Λ఻͑ΔΠϯλϥΫ γϣϯʯΛઃܭ͢Δͱ͍͏ண૝ʹࢸͬͨɽ ਃ੥ऀ͕ߦͬͨ࠷৽ͷݚڀͰ͸ɼೝ஌৺ཧֶʹ͓͚ΔϓϥΠϛϯάޮՌʹண໨͠ɼ൷൑తࢥߟ͕Ͱ ͖Δਓ෺Λ࿈૝͢ΔϫʔυΛ΢Σϒݕࡧ࣌ʹఏࣔ͢Δ͜ͱͰɼ൷൑తͳ৘ใ୳ࡧߦಈΛ༠ൃͰ͖ΔՄ ೳੑ͕ࣔࠦ͞Εͨʢۀ੷ [5]ʣ ɽ͜ͷ͜ͱ͔Βɼೝ஌৺ཧֶ΍ߦಈܦࡁֶͷ஌ݟΛੵۃతʹऔΓࠐΉ͜ ͱͰɼ ʮ৘ใਫ਼ࠪଶ౓Λܹࢗ͠ɼ൷൑త৘ใ୳ࡧʮ༠ൃʯ͢Δ৘ใݕࡧΠϯλϥΫγϣϯʯͱ͍͏ண૝ ʹࢸͬͨɽҰํɼઌͷݚڀͰ͸ɼڧ͍৴೦Λ͍࣋ͬͯΔϢʔβʹ͸ɼ؇΍͔ʹߦಈΛଅ͢φοδత Ξϓϩʔν͸͏·͘ػೳ͠ͳ͍͜ͱ͕ࣔࠦ͞Ε͍ͯΔɽͦͷΑ͏ͳϢʔβʹରͯ͠͸ɼ൷൑త৘ ใ୳ࡧͷඞཁੑΛڧ͘ײͤ͡͞Δඞཁ͕͋Δɽ͜ͷ͜ͱ͔Βɼ ʮ൷൑త৘ใ୳ࡧΛϑΝγϦςʔτ ͢Δʮ໰͍ʯͷର࿩తఏࣔʯͱ͍͏ண૝ʹࢸͬͨɽ ؔ࿈͢Δࠃ಺֎ͷݚڀಈ޲ͱຊݚڀͷҐஔ͚ͮʹ͍ͭͯ ۙ೥ɼ ৘ใݕࡧ ɾ σʔλϚΠχϯάͷ෼໺Ͱ͸ɼ X. YinΒ͕։ൃͨ͠TruthFinder΍J. Pasternack Β͕։ൃͨ͠ Latent Credibility Analysis ͷΑ͏ʹɼ৘ใ৴པੑͷࣗಈධՁΞϧΰϦζϜ͕ݚڀ ։ൃ͞Εͭͭ͋Δ6ɽۙ೥Ͱ͸ɼϑΣΠΫχϡʔεͷݕग़ΞϧΰϦζϜ΋੝Μʹݚڀ։ൃ͞Ε͍ͯ Δ7ɽਤ 1 Ͱࣔͨ͠Α͏ʹɼຊݚڀ՝୊Ͱ͸ɼܭࢉػʹΑΔ৘ใͷ඼࣭ධՁɼ௿඼࣭৘ใͷݕग़Ξ ϧΰϦζϜͷෆ׬શੑΛิ͏ͨΊʹɼ৘ใӾཡϢʔβͷ൷൑త৘ใ୳ࡧεΩϧ΍ଶ౓ͷڧԽΛਤ Δ৘ใΠϯλϥΫγϣϯͷ։ൃʹয఺Λ౰͍ͯͯΔɽ ۙ೥ɼώϡʔϚϯɾίϯϐϡʔλΠϯλϥΫγϣϯͷ෼໺Ͱ͸ɼ΢Σϒ৘ใͷऔࣺબ୒ʹ͔͔ Δೝ஌όΠΞε෼ੳʹؔ͢Δݚڀ͕஫໨ΛूΊ͍ͯΔɽྫ͑͹ɼMicrosoft Research ͷ R. White ͸ɼ΢Σϒݕࡧ࣌ͷ৻ॏͳҙࢥܾఆΛଅਐ͢Δʹ͸ɼݕࡧτϐοΫʹؔͯ͠ࣄલʹ๊͍͍ͯΔό ΠΞεΛऔΓআ͘࢓૊Έ͕ඞཁͱड़΂͍ͯΔ8ɽIBM Research ͷ Liao Β͸ɼࣗ෼ͷझ޲ʹ͋ͬͨ ৘ใΛ༏ઌతʹӾཡͯ͠͠·͏ʮબ୒త஫ҙ܏޲ʯΛ؇࿨͢ΔͨΊͷ৘ใఏࣔख๏Λݚڀ͓ͯ͠ Γɼೝ஌৺ཧֶతཁૉΛߟྀͨ͠৘ใΞΫηεγεςϜͷઃܭͷॏཁੑΛઆ͍͍ͯΔ9ɽຊݚڀ՝ ୊͸ɼ͜ΕΒͷݚڀ͕ࢦఠ͢Δ໰୊Λղܾ͢ΔࢼΈͰ΋͋Δɽ 6J. Pasternack et al.: Latent Credibility Analysis. WWW 2013, pp.1009-1020. 7N. Ruchansky et al.: CSI: A Hybrid Deep Model for Fake News Detection. CIKM 2017, pp.797–806. 8R. White: Beliefs and Biases in Web Search. SIGIR 2013. pp.3-12. 9Q. Liao et al. It Is All About Perspective: An Exploration of Mitigating Selective Exposure with Aspect Indicators. CHI 2015, pp.1439-48.
8. ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̒ ᇶ┙◊✲㸦㹀㸧㸦୍⯡㸧㸴 㸱 ᛂເ⪅ࡢ◊✲㐙⾜⬟ຊཬࡧ◊✲⎔ቃ ᮏḍ࡟ࡣᛂເ⪅㸦◊✲௦⾲⪅ࠊ◊✲ศᢸ⪅㸧ࡢ◊✲ィ⏬ࡢᐇ⾜ྍ⬟ᛶࢆ♧ࡍࡓࡵࠊ  ࡇࢀࡲ࡛ࡢ◊✲άືࠊ  ◊✲⎔ቃ㸦◊ ✲㐙⾜࡟ᚲせ࡞◊✲᪋タ࣭タഛ࣭◊✲㈨ᩱ➼ࢆྵࡴ㸧࡟ࡘ࠸࡚㸰㡫௨ෆ࡛グ㏙ࡍࡿࡇ࡜ࠋ ࠕ  ࡇࢀࡲ࡛ࡢ◊✲άືࠖࡢグ㏙࡟ࡣࠊ◊✲άືࢆ୰᩿ࡋ࡚࠸ࡓᮇ㛫ࡀ࠶ࡿሙྜ࡟ࡣࡑࡢㄝ᫂࡞࡝ࢆྵࡵ࡚ࡶࡼ࠸ࠋ ͤ␃ព஦㡯 1. ◊✲ᴗ⦼㸦ㄽᩥࠊⴭ᭩ࠊ⏘ᴗ㈈⏘ᶒࠊᣍᚅㅮ₇➼㸧ࡣࠊ⥙⨶ⓗ࡟グ㍕ࡍࡿࡢ࡛ࡣ࡞ࡃࠊ ᮏ◊✲ィ⏬ࡢᐇ⾜ྍ⬟ᛶࢆㄝ᫂ࡍࡿୖ࡛ࠊࡑࡢ᰿ᣐ࡜࡞ࡿᩥ⊩➼ࡢ୺せ࡞ࡶࡢࢆ㐺ᐅグ ㍕ࡍࡿࡇ࡜ࠋ 2. ◊✲ᴗ⦼ࡢグ㏙࡟ᙜࡓࡗ࡚ࡣࠊᙜヱ◊✲ᴗ⦼ࢆྠᐃࡍࡿ࡟༑ศ࡞᝟ሗࢆグ㍕ࡍࡿࡇ࡜ࠋ ౛࡜ࡋ࡚ࠊᏛ⾡ㄽᩥࡢሙྜࡣㄽᩥྡࠊⴭ⪅ྡࠊᥖ㍕ㄅྡࠊᕳྕࡸ㡫➼ࠊⓎ⾲ᖺ㸦すᬺ㸧ࠊ ⴭ᭩ࡢሙྜࡣࡑࡢ᭩ㄅ᝟ሗࠊ࡞࡝ࠋ 3. ㄽᩥࡣࠊ᪤࡟ᥖ㍕ࡉࢀ࡚࠸ࡿࡶࡢཪࡣᥖ㍕ࡀ☜ᐃࡋ࡚࠸ࡿࡶࡢ࡟㝈ࡗ࡚グ㍕ࡍࡿࡇ࡜ࠋ 4. ᮏ␃ព஦㡯㸦ᩳయࡢᩥ᭩㸧ࡣࠊ◊✲ィ⏬ㄪ᭩ࡢసᡂ᫬࡟ࡣ๐㝖ࡍࡿࡇ࡜ࠋ ͜Ε·Ͱͷݚڀ׆ಈʢݚڀ߲໨ͱରԠ͚ͮͯهड़ʣ ͜Ε·Ͱਃ੥ऀ͸ɼ৘ใݕࡧɾώϡʔϚϯίϯϐϡʔλΠϯλϥΫγϣϯͷ෼໺Ͱɼ΢Σϒ৘ใ ͷ৴པੑ෼ੳΞϧΰϦζϜ [12]ɼ৴པੑ൑அΛߦ͏ͨΊͷσʔλϚΠχϯάٕज़ʢۀ੷ [10, 9]ʣ ɼ ৴པੑ൑அͷͨΊͷݕࡧΠϯλϑΣʔεʢۀ੷ [11, 9]ʣ ɼ৴པੑ൑அΛߦ͏Ϣʔβͷߦಈɾଶ౓ ಛੑʹؔ͢Δ෼ੳʢۀ੷ [7, 8]ʣʹؔ͢ΔݚڀΛߦ͖ͬͯͨɽҎԼͰ͸ɼ͜Ε·Ͱͷݚڀ׆ಈͷ͏ ͪɼຊݚڀ՝୊ʹؔ࿈͢Δ΋ͷΛऔΓ্͛Δɽͳ͓ɼݚڀ୅දऀ͸ɼ2013 ೥͔Β 2015 ೥ʹ͔͚ ͯɼژ౎େֶϦαʔνɾΞυϛχετϨʔλͱͯ͠ݚڀϚωδϝϯτͷઐ໳ۀ຿ʹैࣄ͍ͯͨ͠ ͨΊɼͦͷظؒ͸ݚڀ׆ಈΛதஅ͍ͯͨ͠ɽ ݚڀ߲໨ 1: ࣗ਎ͷ৘ใ୳ࡧߦಈΛ಺লɾվળ͢ΔͨΊͷ΢Σϒϒϥ΢β ݚڀ୅දऀ͸ɼϢʔβͷ൷൑తࢥߟଶ౓ͱ΢Σϒ୳ࡧߦಈͱͷؔ܎෼ੳɼ৴ጪੑͷߴ͍΢Σϒ ৘ใͷ֫ಘࢧԉͷͨΊͷϢʔβߦಈͷཧղʹؔ͢ΔݚڀΛߦ͖ͬͯͨʢۀ੷ [11, 9, 7, 8, 4, 1]ʣ ɽ ͦͷ੒ՌͷҰ෦͸ɼCHIɼCIKMɼHT ͳͲͷτοϓϨϕϧࠃࡍձٞʹ࠾୒͞Ε͍ͯΔɽຊݚڀ߲ ໨Ͱ͸ɼ͜Ε·ͰͷݚڀͰಘͨ൷൑త৘ใ୳ࡧʹඞཁͱͳΔεΩϧ΍ଶ౓ʹؔ͢Δ஌ݟΛ΋ͱʹɼ ࣗ਎ͷ൷൑త৘ใ୳ࡧߦಈΛ಺লɾվળ͢ΔͨΊͷΠϯλϑΣʔεʹ͍ͭͯݚڀΛߦ͏ɽ ݚڀ߲໨ 2: φοδΛ༻͍ͨɼ৘ใਫ਼ࠪଶ౓Λܹࢗ͢Δݕࡧ݁ՌαϚϦͷੜ੒ʢ҉໧తઆಘʣ ݚڀ୅දऀ͸ɼ΢Σϒ৘ใͷ৴པੑΛҙ͍ࣝͯ͠ͳ͍ݕࡧϢʔβʹରͯ͠஫ҙਂ͍৘ใݕࡧΛ ༠ൃ͢Δ৘ใΠϯλϥΫγϣϯΛ։ൃ͖ͯͨ͠ʢۀ੷ [5, 6, 2]ʣ ɽೝ஌৺ཧֶͷ஌ݟͱ৘ใݕࡧͷ ஌ݟΛ૊Έ߹Θͤͨ৽͍͠ݕࡧΠϯλϥΫγϣϯʹؔ͢Δݚڀ੒Ռ͸ɼWebDB ϑΥʔϥϜ 2019 ࠷༏ल࿦จ৆ʢۀ੷ [5]ʣ΍ ACM/IEEE JCDL2020 Vannevar Best Paper Awardʢۀ੷ [2]ʣΛ त৆͢ΔͳͲɼߴ͍ධՁΛಘͨɽຊݚڀ߲໨Ͱ͸ɼಛఆͷݕࡧτϐοΫͰ͸ͳ͘෯޿͍τϐοΫ ͷ৘ใݕࡧʹద༻Ͱ͖ΔΑ͏ɼ͜Ε·Ͱ։ൃ͖ͯͨ͠ߦಈ༠ൃͷͨΊͷ৘ใΠϯλϥΫγϣϯٕ ज़Λൃలͤ͞ɼߦಈܦࡁֶ౳ͷ஌ݟ΋औΓೖΕͭͭɼ൷൑త৘ใ୳ࡧΛʮ༠ൃʯ͢Δݕࡧ݁Ռα ϚϦͷੜ੒ख๏ͷཱ֬Λ໨ࢦ͢ɽ ݚڀ߲໨ 3: ൷൑త৘ใ୳ࡧΛϑΝγϦςʔτ͢Δʮ໰͍ʯͷର࿩తఏࣔʢ໌ࣔతઆಘʣ ݚڀ୅දऀ͸ɼQA αΠτʹ͓͚Δ࣭໰Ԡ౴ΠϯλϥΫγϣϯͷ෼ੳʹऔΓ૊ΜͰ͓Γɼͦͷ աఔͰɼQA αΠτʹ͸Ϣʔβʹؾ͖ͮΛ༩͑ΔλΠϓͷ໰͍͕ଘࡏ͢Δ͜ͱΛ໌Β͔ʹ͠ɼຊ ݚڀ߲໨Λண૝͢Δʹࢸͬͨʢۀ੷ [3]ʣ ɽ͜Ε·Ͱɼݚڀ୅දऀ͸ɼࣗવݴޠॲཧٕज़/৘ใݕࡧ ٕज़Λ׆༻͠ɼ΢Σϒ৘ใͷ৴པੑ൑அࢧԉΛߦ͏ͨΊͷ৘ใΛϚΠχϯά/ݕࡧ͢ΔΞϧΰϦζ Ϝͷ։ൃΛߦ͖ͬͯͨ͜ͱ΋͋Γɼࣗવݴޠܗࣜͷʮ໰͍ʯͷϚΠχϯά΍ϚΠχϯά݁Ռͷ࠷ దͳఏࣔख๏ͷઃܭʹඞཁͳεΩϧ͸े෼ʹ༗͍ͯ͠Δʢۀ੷ [10]ʣ ɽ ݚڀ؀ڥ ݚڀ୅දऀ͸ɼຊݚڀ՝୊ʹؔ͢Δٞ࿦΍࣮૷ɼ࣮ݧΛߦ͏ͨΊͷे෼ͳεϖʔεΛ༗͍ͯ͠ Δɽݚڀ਱ߦͷͨΊͷઃඋʹ͍ͭͯ͸ɼσʔλ෼ੳɾݚڀ࣮ࣨݧ༻ͷܭࢉػΛ਺୆༗͍ͯ͠Δ͕ɼ Ϣʔβ࣮ݧ΍σʔλղੳΛΑΓ҆ఆతͳ؀ڥͰߦ͏ͨΊʹɼຊݚڀܦඅʹͯΞϓϦέʔγϣϯαʔ όͱ෼ੳ༻αʔόΛ 1 ୆ͣͭߪೖ͢Δ༧ఆͰ͋Δɽݚڀ߲໨ 2 ͓Αͼ 3 Ͱ͸ϨϏϡʔαΠτͷจॻ ίʔύε΍ QA ίʔύεΛ࢖ͬͨݚڀΛߦ͏༧ఆͰ͋Δ͕ɼ৘ใݯͱͳΔσʔλͷҰ෦ʢYahoo! ஌ܙାʣ͸طʹ༗͓ͯ͠Γɼਝ଎ʹݚڀ՝୊ʹऔΓ૊Ή͜ͱ͕ՄೳͰ͋Δɽ
9. ʲ̏ɹԠืऀͷݚڀ਱ߦೳྗٴͼݚڀ؀ڥʢ͖ͭͮʣ ʳ ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̓ ຊ՝୊Ͱ͸ɼ৘ใ୳ࡧߦಈͷ಺লΛଅ͢ΠϯλϑΣʔεʢݚڀ߲໨ 1ʣ΍൷൑త৘ใ୳ࡧΛʮ༠ ൃʯ͢Δݕࡧ݁ՌαϚϦੜ੒ʢݚڀ߲໨ 2ʣͷݚڀ։ൃΛߦ͏ͨΊʹɼߦಈܦࡁֶ΍ೝ஌৺ཧֶɼ ڭҭ޻ֶͷ஌ݟΛ׆༻͢Δɽຊ՝୊͸ݚڀ୅දऀͷΈ͔Βߏ੒͞ΕΔνʔϜͰ࣮ߦ༧ఆͰ͋Δ͕ɼ ݚڀ୅දऀ͕ॴଐ͢Δ੩Ԭେֶ৘ใֶ෦ʹ͸ɼߦಈܦࡁֶ΍ڭҭ޻ֶɼೝ஌৺ཧֶͷݚڀऀ͕͍ ΔɽͦͷͨΊɼͦΕΒݚڀऀ͔ΒఆظతʹॿݴΛ΋Β͍ͳ͕Βɼݚڀ՝୊Λୡ੒Λ໨ࢦ͢ɽ ؔ࿈͢Δݚڀۀ੷ [1] S. Pothirattanachaikul, T. Yamamoto, Y. Yamamoto and M. Yoshikawa: Analyzing the Effects of ”People also ask” on Search Behaviors and Beliefs. Procs. of the 31st ACM Conference on Hypertext and Social Media (HT 2020), pp.101-110, 2020 (ࠪಡ͋Γ). [2] Y. Yamamoto and T. Yamamoto: Personalization Finder: A Search Interface for Identi- fying and Self-controlling Web Search Personalization. Procs. of the 20th ACM/IEEE on Joint Conference on Digital Libraries (JCDL 2020), pp.37-46, 2020 (ࠪಡ͋Γ). [3] ᴡ౻࢙໌, ࢁຊ༞ี:ʮQA αΠτʹ͓͚Δ࣭໰Ԡ౴ʹண໨ͨ͠ؾ͖ͮΛଅ͢໰͍͔͚ͷ෼ੳʯ, ୈ 12 ճσʔλ޻ֶͱ৘ใϚωδϝϯτʹؔ͢ΔϑΥʔϥϜʢDEIM2020ʣ, 2020ʢࠪಡͳ͠ʣ. [4] S. Pothirattanachaikul, T. Yamamoto, Y. Yamamoto and M. Yoshikawa: Analyzing the Effects of Document’s Opinion and Credibility on Search Behaviors and Belief Dynamics. Procs. of the 28th ACM Conference on Information and Knowledge Management (CIKM 2019), pp.1653-1662, 2019 (ࠪಡ͋Γ). [5] ࢁຊ༞ี, ࢁຊַ༸: ʮ൷൑తͳ΢ΣϒݕࡧΛଅਐ͢ΔΫΤϦϓϥΠϛϯάʯ, ৘ใॲཧֶձ ࿦จࢽ: σʔλϕʔε (TOD80), Vol.12, No.1, pp.38-52, 2019ʢࠪಡ͋Γʣ. [6] F. Saito, Y. Shoji and Y. Yamamoto: Highlighting Weasel Sentences for Promoting Critical Information Seeking on the Web. Procs. of the 20th International Conference on Web Information Systems Engineering (WISE 2019), pp.424-440, 2019 (ࠪಡ͋Γ). [7] T. Yamamoto, Y. Yamamoto and S. Fujita: Exploring People’s Attitudes and Behaviors toward Careful Information Seeking in Web Search. Procs. of the 27th ACM Conference on Information and Knowledge Management (CIKM 2018), pp.963-972, 2018 (ࠪಡ͋Γ). [8] Y. Yamamoto, T. Yamamoto, H. Ohshima, and K. Hiroshi: Web Access Literacy Scale to Evaluate How Critically Users Can Browse and Search for Web Information. Procs. of the 10th ACM Conference on Web Science (WebSci 2018), pp.97-106, 2018 (ࠪಡ͋Γ). [9] Y. Yamamoto and S. Shimada: Can Disputed Topic Suggestion Enhance User Consider- ation of Information Credibility in Web Search?. Procs. of the 27th ACM Conference on Hypertext and Social Media (HT 2016), pp.169-177, 2016 (ࠪಡ͋Γ). [10] Y. Yamamoto: Disputed sentence suggestion towards credibility-oriented web search. Procs. of the 14th Asia-Pacific international conference on Web Technologies and Applications (APWeb 2012), pp.34-45, 2012 (Best paper first-runner up) (ࠪಡ͋Γ) . [11] Y. Yamamoto and K. Tanaka: Enhancing credibility judgment of web search results. Procs. of the SIGCHI Conference on Human Factors in Computing Systems (CHI 2011), pp.1235- 1244, 2011 (ࠪಡ͋Γ) . [12] ࢁຊ༞ี, ాதࠀݾ: ʮσʔλରؒͷαϙʔτؔ܎෼ੳʹجͮ͘ Web ৘ใͷ৴ጪੑධՁʯ, ৘ ใॲཧֶձ࿦จࢽ: σʔλϕʔε (TOD46), Vol.3, No.2, pp.61-79, 2010 (ࠪಡ͋Γ).
10. ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̔ ᇶ┙◊✲㸦㹀㸧㸦୍⯡㸧㸶 㸲 ேᶒࡢಖㆤཬࡧἲ௧➼ࡢ㑂Ᏺ࡬ࡢᑐᛂ㸦බເせ㡿㸲㡫ཧ↷㸧 ᮏḍ࡟ࡣࠊᮏ◊✲ࢆ㐙⾜ࡍࡿ࡟ᙜࡓࡗ࡚ࠊ┦ᡭ᪉ࡢྠព࣭༠ຊࢆᚲせ࡜ࡍࡿ◊✲ࠊಶே᝟ሗࡢྲྀᢅ࠸ࡢ㓄៖ࢆᚲせ࡜ࡍࡿ◊✲ࠊ ⏕࿨೔⌮࣭Ᏻ඲ᑐ⟇࡟ᑐࡍࡿྲྀ⤌ࢆᚲせ࡜ࡍࡿ◊✲࡞࡝ᣦ㔪࣭ἲ௧➼㸦ᅜ㝿ඹྠ◊✲ࢆ⾜࠺ᅜ࣭ᆅᇦࡢᣦ㔪࣭ἲ௧➼ࢆྵࡴ㸧 ࡟ᇶ࡙ࡃᡭ⥆ࡀᚲせ࡞◊✲ࡀྵࡲࢀ࡚࠸ࡿሙྜࠊㅮࡌࡿᑐ⟇࡜ᥐ⨨ࢆࠊ㸯㡫௨ෆ࡛グ㏙ࡍࡿࡇ࡜ࠋ ಶே᝟ሗࢆక࠺࢔ࣥࢣ࣮ࢺㄪᰝ࣭࢖ࣥࢱࣅ࣮ࣗㄪᰝ࣭⾜ືㄪᰝ㸦ಶேᒚṔ࣭ᫎീࢆྵࡴ㸧ࠊᥦ౪ࢆཷࡅࡓヨᩱࡢ౑⏝ࠊࣄࢺ 㑇ఏᏊゎᯒ◊✲ࠊ㑇ఏᏊ⤌᥮࠼ᐇ㦂ࠊື≀ᐇ㦂࡞࡝ࠊ◊✲ᶵ㛵ෆእࡢ೔⌮ጤဨ఍➼࡟࠾ࡅࡿᢎㄆᡭ⥆ࡀᚲせ࡜࡞ࡿㄪᰝ࣭◊ ✲࣭ᐇ㦂࡞࡝ࡀᑐ㇟࡜࡞ࡾࡲࡍࠋ ヱᙜࡋ࡞࠸ሙྜ࡟ࡣࠊࡑࡢ᪨グ㏙ࡍࡿࡇ࡜ࠋ ຊݚڀ՝୊Ͱ͸ɼ՝୊਱ߦͷͨΊʹ΢ΣϒจॻʢQA αΠτɼϨϏϡʔαΠτ౳ͷ΢Σϒϖʔ δʣΛऩूɾ෼ੳ͢Δɽ͜ΕΒͷσʔλ͸ɼݪଇͱͯ͠ɼ΢Σϒ্ʹެ։͞Ε͍ͯΔΦʔϓϯσʔ λΛऩूର৅ͱ͠ɼݸਓ৘ใؚ͕·ΕΔ΋ͷ͸࢖༻͠ͳ͍ɽ·ͨɼQA αΠτσʔλͷΑ͏ʹɼα ΠτϙϦγʔʹΑͬͯσʔλͷऩूɾ෼ੳ੍͕ݶ͞Ε͍ͯΔ΋ͷʹ͍ͭͯ͸ɼผ్Πϯλʔωο ταʔϏεϓϩόΠμͱܖ໿Λߦ͍ɼσʔλͷೖखɾ෼ੳΛߦ͏ɽݸਓ৘ใؚ͕·ΕΔσʔλΛ ࢖༻͢Δ৔߹͸ɼݸਓ৘ใอޢؔ܎ͷ๏ྩʹै͏ɽ ্هσʔλʹՃ͑ɼຊ՝୊Ͱ͸ɼݚڀ࣮ࣨݧ΍Ϋϥ΢υιʔγϯάΛ௨ͯ͡ɼϢʔβͷ΢Σϒ ݕࡧϩά΍ࢥߟಛੑɼؔ৺ࣄ߲ʹؔ͢ΔΞϯέʔτσʔλͷऩूΛߦ͏ɽ͜ΕΒσʔλͷऔΓѻ ͍ʹ͍ͭͯ͸ɼ ʮ੩Ԭେֶʹ͓͚Δݸਓ৘ใอޢϙϦγʔʯʹج͖ͮɼ࠷େݶͷ஫ҙΛ෷͏ɽϩά σʔλ͸ɼઐ༻ͷ୺຤Ҏ֎͔Β͸ΞΫηεͰ͖ͳ͍αʔόͰ؅ཧɾอଘΛߦ͏ɽͦͷࡍɼݸਓ͕ ಛఆͰ͖ͳ͍Α͏ɼಗ໊ԽΛߦ্ͬͨͰσʔλΛอଘ͢Δɽ ຊ՝୊Ͱ͸ݸਓΛಛఆ͢Δ৘ใͷऩू͸ߦΘͳ͍͕ɼݸਓͷࢥߟಛੑ΍ݕࡧೳྗʹؔ͢Δ৘ใ ͷऩूΛؚΉɽͦͷͨΊɼඞཁʹԠͯ͡େֶ಺ͷྙཧҕһձͰͷঝೝΛड͚ͨޙɼௐࠪɾ࣮ݧΛ ࣮ࢪ͢Δɽඃݧऀ͔Βݸਓ৘ใΛಘΔࡍʹ͸ɼ໨త΍ެ։ͷൣғɼ؅ཧํ๏ʹ͍ͭͯࣄલʹઆ໌ Λߦ͏ɽ·ͨɼඃݧऀ͔Β໰͍߹Θ͕ͤ͋ͬͨ৔߹ʹ͸ɼ଎΍͔ʹ͜ΕʹԠ͡ɼσʔλ࡟আʹ΋ Ԡ͡Δɽ࣮ݧڠྗँۚ౳ͷࢧ෷͍ʹඞཁͳݸਓ৘ใʹ͍ͭͯ͸ɼࣄ຿खଓ͖͕ऴྃ࣍ୈɼഁغΛ ߦ͏ɽ Ξϯέʔτௐ͓ࠪΑͼϢʔβ࣮ݧ౳Ͱ࣮ݧڠྗऀʹใुΛࢧ෷͏ࡍʹ͸ɼෆ౰ʹ௿͍ใुֹʹ ͳΒͳ͍Α͏ɼ࠷௿௞ֹۚ΍ಉ༷ͷ࣮ݧͷใुֹ౳ΛࢀߟʹใुֹΛઃఆ͢Δɽ
11. ج൫ݚڀʢ̗ʣ ʢҰൠʣ ̕ ᇶ┙◊✲㸦㹀㸧㸦୍⯡㸧㸷 㸳 ◊✲ィ⏬᭱⤊ᖺᗘ๓ᖺᗘᛂເࢆ⾜࠺ሙྜࡢグ㏙஦㡯㸦ヱᙜ⪅ࡣᚲࡎグ㏙ࡍࡿࡇ࡜㸦බເせ㡿㸰㸳㡫ཧ↷㸧㸧 ᮏḍ࡟ࡣࠊᮏ◊✲ࡢ◊✲௦⾲⪅ࡀ⾜ࡗ࡚࠸ࡿࠊ௧࿴㸱  ᖺᗘࡀ᭱⤊ᖺᗘ࡟ᙜࡓࡿ⥅⥆◊✲ㄢ㢟ࡢᙜึ◊✲ィ⏬ࠊࡑࡢ◊ ✲࡟ࡼࡗ࡚ᚓࡽࢀࡓ᪂ࡓ࡞▱ぢ➼ࡢ◊✲ᡂᯝࢆグ㏙ࡍࡿ࡜࡜ࡶ࡟ࠊᙜヱ◊✲ࡢ㐍ᒎࢆ㋃ࡲ࠼ࠊᮏ◊✲ࢆ๓ᖺᗘᛂເࡍࡿ⌮⏤ 㸦◊✲ࡢᒎ㛤≧ἣࠊ⤒㈝ࡢᚲせᛶ➼㸧ࢆ㸯㡫௨ෆ࡛グ㏙ࡍࡿࡇ࡜ࠋ ヱᙜࡋ࡞࠸ሙྜࡣグ㏙ḍࢆ๐㝖ࡍࡿࡇ࡜࡞ࡃࠊ✵ḍࡢࡲࡲᥦฟࡍࡿࡇ࡜ࠋ ◊✲✀┠ྡ ㄢ㢟␒ྕ ◊✲ㄢ㢟ྡ ◊✲ᮇ㛫    ᖹᡂ ᖺ ᗘ㹼௧࿴ 㸱ᖺᗘ ᙜึ◊✲ィ⏬ཬࡧ◊✲ᡂᯝ ๓ᖺᗘᛂເࡍࡿ⌮⏤ ݚڀछ໨໊ ՝୊൪߸ ݚڀ՝୊໊ ݚڀظؒ ฏ੒ ೥౓ ʙྩ࿨ 3 ೥౓ ౰ॳݚڀܭըٴͼݚڀ੒Ռ લ೥౓Ԡื͢Δཧ༝