The 26th Annual Conference of the Japanese Society for Artificial Intelligence, 2012 4E1-OS-15-4 ᪂Ϊঠ஦΅⁄‫⁉•⁜⁆‷‫‬ⁱ‫‮‬Ύ ࿇ᮇᕷሙືྥ΅ศᯒ Analysis of Long-term Market Trend by Text-Mining of News Articles ‫܍‬ᮏ ଇஂ*1 ࿴Ἠ ₩*1 Takahisa Kuramoto Kiyoshi Izumi ୰ᔱ ၨᾈ*2 Akihiro Nakashima ྜྷᮧ ᚸ*1 Shinobu Yoshimura ᯇ஭ ‫ܘ‬஬൫*3 Tohgoroh Matsui *1ូ ▼⏣ ᬛஓ*2 Tomonari Ishida ྜྷ⏣ ⛱*4 Minoru Yoshida ୰ᕝ ࣡ᚿ*4 Hiroshi Nakagawa ᮾி኱Ꮫ኱Ꮫ၄ូ ᕤᏛǄ◊"⛉ School of Engineering, The University of Tokyo *2 ญᮧ੷ๆᰴᘧ఍♫ᬌ Nomura Securities Co., Ltd *3 ୰ൺ኱Ꮫូ ⏕࿨೺ᗣ⛉Ꮫൺ College of Life and Health Sciences , Chubu University *4 ᮾி኱Ꮫូ ᝟ሗᇶ┙‹ⁱ‽⁺ Information Technology Center, The University of Tokyo Abstract: In this study, we developed a new method of the long-term market analysis by using text-mining of news articles. Using our method, we conducted extrapolation tests to predict stock price averages by 19 industry and two market averages, TOPIX and Nikkei225 for about 10 years. As a result, 8 sectors in 21 sectors (about 40%) showed over about 60% accuracy, and 15 sectors in 21 sectors (over 70%) showed over about 55% accuracy. We also developed a web system of financial text-mining based on our method for financial professionals. 1. `Ῐ Ύ άືΎᙺWῤ᝟ሗ‒ᥦ౪Ῑ⃺ Ȃ῭΅ኚື`Ȃ῭ᣦᶆ ᰴ౯Ύ௦ࢢ῕‌ᩘ್⁅⁺‽⃸⁉⁣⁺ ‷ ᪂ΪΎ௦ࢢ῕‌⁄‫⁅⁆‷‫‬⁺‽ῪῩ⃸Ѯ኱Ὺ᝟ሗῌ┦஫Ύస ⏝ῗྜΰῦỴᐃ῕‌῟ ჰɪᙧῪᣲື‒♧Ῑ⃺῝‌Ύᑐῗῦᢞ ଘᐙ`⁄‭⁉ ⁩ศᯒ ⁓‟ⁱ‾ ⁱ‽⁩‸ศᯒ΅ ῆῪศᯒᡭἲ ‒ቁ౑ῗῦᢞଘ‒ࢌΰῦῄῌ⃸῝‌ῧ ࿇ᮇ΅ኚື΅ண `ჰ ᖖΎᅔႤῪ΅ῌ⌧≧ῧῂ⃺ΐ‌`᝟ሗฎ΅Ѯ኱῕⃸᝟ሗ࿛΅࿥ ಀᛶ΅ࣽ႗῕ΎୣᅉῗῦῊ ⃸῝΅᝟ሗ඲ῦ‒▐᫬Ύॾจῗῦപ ษῪᢞଘ‒ࢌῆΐῨ`୙ྍЏΎಶῄ⃺࿇ᮇ΅ኚືΎῊῄῦ`⃸ኚື ‒ỴᐃῙ᝟ሗ ῕ ΎѮ኱ΎῪ ⃸῝΅࿥ಀᛶ ୙᫂░ΎῪ῟ ⃸ᢞଘุ᩿`┈ᾫႤῗ῏Ὺ⃺ ΐ‌῾ῧ΅◊"ῧ`⁉⁣⁺⁧⁩⁋⁁⁆⁭⁺‭>዇ @ ഷఏ ⓗ⁕⁫‫⁝⁧‮‬ⁱ‫>‮‬ఀᗞ @Ῠῄΰ῟ᶵᲔᏛ͖΅ᡭἲΎ ΰῦȂ ῭΅ኚື‒ҜືⓗΎศᯒῙΐῨ‒┠ᣦῗ῟ ΅ῌᏑᅾῙ⃺ῗΉ ῗ⃸ᩘ್᝟ሗ΅ॾᯒȆᯝ΅῿ῧ࿇ᮇⓗῪᢞଘุ᩿‒ୗῙΐῨ`Ⴄ ῗῄ⃺῝΅⌮⏤Ῠῗῦ⃸ே࿛ῌॾจ‒ࢌῈῪῄΏῩࣽ႗Ὺ᭱പ໬ῌ ⏕ῘΐῨ ⃸ᩘ್እ΅᝟ሗῌḞؐῗῦῗ῾ΰῦῄΐῨῌᣲῒ ‌ ⃺῝ΐῧ⁄‫⁆‷‫‬᝟ሗ‒⏝ῄῦȂ῭΅ኚື‒ศᯒῙ◊"ῌಶᖺ ⌧‌ῦ῍῟⃺࿴Ἠ ` &35 ἲῨῄῆᡭἲ‒࿔Ⓨῗ⃸⁄‫ॾ‒⁆‷‫‬ᯒ ῙΐῨῧȂ῭ືྥ΅ศᯒ‒ࢌΰῦῄ>࿴Ἠ @⃺⁄‫⏝‒⁆‷‫‬ ῄ฼ⅬῨῗῦ`ॾจῌᐜ᫆ῧῂΐῨ⃸ᩘ್Ύྵ῾‌Ὺῄ᝟ሗ ศᯒῧ῍ΐῨ΅ Ⅼῌᣲῒ ‌⃺ ᮏ◊"ῧ`⁄‫⁆‷‫‬᝟ሗ‒⏝ῄ⃸ᢞଘᐙΉ ΅ोᮃῌ዇ῄ࿇ᮇ ⓗῪȂ῭ືྥ΅ศᯒ‒ࢌῆ⃺῕ ΎѮ኱Ὺ᝟ሗ΅୰Ή ᢞଘΎ᭷ ┈Ὺ᝟ሗ‒ᢳฟῗ⃸῝΅᝟ሗ࿛΅࿥ೳᛶ‒᫂☜ΎῙΐῨῧᢞଘ 2. ศᯒᡭἲ ᬌᮏ◊"΅ෆᐜ`ญᮧ੷ๆᰴᘧ఍♫΅බᘧ॒ॾ‒♧Ῑ ΅ῧ` ῂ ῾Ί–⃺ូ ೳȑඛ៖ូ ࿴Ἠ₩⃸ᮾி኱Ꮫ኱Ꮫ၄ᕤᏛǄ◊"⛉‵‷⁄⁞๰ᡂᏛᑓ ᨷ⃸Ᾰ៍៍៏៉។្៑្ូ ᩥி༊ᮏඃ ៓៉៏៉៍⃸ូ ᠅᠖᠑᠉᠅ៜ᠏᠕᠏៊᠐៊᠑៉᠐᠋᠇᠕᠋៊៽៿៊᠆᠌ូ ᮏ◊"ῧ`࿴Ἠ ῌᥦၐῙ CPR ἲ‒⏝ῄῦศᯒ‒ࢌῆ⃺CPR ἲῨ`ඹୣॾᯒ⃸୺ᡂศศᯒ⃸ᅇᖐศᯒ΅୕ẁၥΉ Ὺศᯒᡭ ἲῧῂ[࿴Ἠ 2011]⃺ᚑ᮶΅ CPR ἲῧධຊ⁄‫⁆‷‫‬Ῠῗῦῄ῟᪥ ᮏຏࢌ΅ฐࠕȂ῭᭶ሗ`ẚుⓗᙧᘧ΅Ỵ῾ΰ῟ᩥ᭩ῧῂ῟ ⃸ ჰᖖΎᢅῄ Ῑῄ ΅ῧῂΰ῟⃺ᮏ◊"ῧ`᪂Ϊঠ஦Ῠῄῆᙧᘧ ῌฐࠕȂ῭᭶ሗ ᐃ῾ΰῦῄῪῄᩥf‒⏝ῄ⃸ ᗈúῧ࿇ᮇ ⓗῪศᯒῌྍЏῨῪ ῆΎ CPR ἲ‒ᣑᙇῗ῟⃺ 2.1 ඹୣॾᯒ ᮏ◊"ῧ⏝ῄ῟⁄‫⁆‷‫‬᝟ሗ`᪥ᮏȂ῭᪂Ϊῧῂ⃺Ȃ῭ǜῧ ῂ῟ ⃸Ȃ῭΅ኚື‒ỴᐃῙोᅉῌᥖ౅῕‌ῦῄῨͳῈ ‌ ⃺ ῾Ὶᆅ᪉ჳ‒၆῏඲ঠ஦Ύᑐῗῦ ChaSen[ChaS]‒⏝ῄῦᙧែǢ ॾᯒ‒ࢌῄ⃸ື৒⁹ྡ৒⁹ᙧᐜ৒‒ᢳฟῙ⃺῝ῗῦྠ୍ᩥ୰Ύၴ ᥋ῗῦฟ⌧ῗ῟༢ਂ΅ῆῡ⃸ᑡῪ῏Ῠ ୍᪉Ύ᪥Ȃ‵※⁺⁧‷>ូ ᪥ Ȃ@Ύ཰༃῕‌ῦῄȂ῭ᑓ࿋⏝ਂῌྵ῾‌ǼྜΊ΅῿‒ᩘῈ ୖῒ⃺᪥Ȃ‵※⁺⁧‷Ῠ`᪥ᮏȂ῭⁅‶‽⁩ ⁅‡†ῌබ࿔ῗῦ ῄ᪂Ϊঠ஦᳨Ǥ΅῟ ΅಑᭩ῧῂ ⃸ǉ1 ୓3 ༓ਂῌ཰༃῕‌ῦ ῄ⃺༢ਂ῔Ῠ΅ฟ⌧ᆆᗘ‒ᩘῈୖῒ ⃸ၴ᥋ῗ῟ඹୣ࿥ಀ ΅ฟ⌧ᆆᗘ‒ᩘῈୖῒΐῨῧȂ῭ືྥΎ࿥Ῑ᝟ሗ‒ῆ῾῏ᢳฟ ῧ ῍⃸ॾจ ᐜ᫆Ύ Ὺ ῨͳῈ ‌ ⃺ᚑ᮶΅ CPR ἲῧ` KeyGraph †⁩′⁨‸⁞[኱⃝ 2006]‒⏝ῄῦฌोਂ‒ồ ῦῄῌ⃸ ᮏ◊"`᪂Ϊঠ஦‒⏝ῄ῟࿇ᮇண ῧῂ῟ ⃸᪥Ȃ‵※⁺⁧‷ ΅ਂῨ΅ඹୣ‒ῨΐῨῧɉ̦ⓗῪ᝟ሗ‒ᢳฟῗ῟⃺ΐ΅ᩘῈୖῒ ‒ ⁴᭶࿛΅ঠ஦ῧˎ ಹῗࢌῄ⃸࿸್௨ୖῪ ฟ⌧⃸࿸್ᮍ‶Ὺ ୙ฟ⌧Ῠῗ⃸ฟ⌧⁏‽⁺ⁱ‒ᐃ̿ῗ῟⃺ῪῊ⃸࿸್` Ῠῗ῟⃺1 ⁴᭶࿛Ύฟ⌧ῗ῟ඹୣ࿥ಀ΅ῆῡᑡῪ῏Ῠ ୍᪉Ύ᪥Ȃ‵※⁺⁧‷ -1- The 26th Annual Conference of the Japanese Society for Artificial Intelligence, 2012 2.2 ୺ᡂศศᯒ 2.1 ÷΅ඹୣॾᯒ‒ഓཤ 3 ᖺ࿛⃴36 ⁴᭶⃵΅᪂Ϊঠ஦Ύᑐῗῦࢌ ῄ⃸࿸್௨ୖῪ ῰ 1⃸࿸್ᮍ‶Ὺ ῰ 0 Ῠῗ⃸1 ⁴᭶῔Ῠ΅༢ਂ΅ ฟ⌧⁏‽⁺ⁱ‒Ȇྜῗ῟ࢌิ‒సᡂῙ⃺ΐ΅᫬⃸ছɾᮇ࿛ῧᑡῪ ῏Ῠ ୍ᅇ`ฟ⌧ῗ῟༢ਂᩘ`ǉ 2 ༓ਂῧῂ ⃸36 ࢌ 2000 ิ΅ࢌ ิῌసᡂ῕‌῟ΐῨΎῪ⃺ΐ΅ࢌิΎᑐῗῦ୺ᡂศศᯒ‒ࢌῄ⃸ ྛ᭶ 15 ಶ΅୺ᡂศῧঠ஦‒ো౯ῗ῟⃺ῙῪ‫‏‬ῡ⃸1 ⁴᭶࿛΅᪂Ϊ ঠ஦‒ 15 ḟඖ΅⁗‭⁆⁩ῧো౯ῗ⃸῝΅⁗‭⁆⁩‒ȆྜῙΐῨῧ ᪂Ϊঠ஦΅≉ᚩฎ΅᫬Ǆิ⁅⁺‽ῌᚓ ‌῟ΐῨΎῪ⃺ 2.3 ᅇᖐศᯒ ᰴ౯⁅⁺‽` 12085$ ΅ ᴗ✀⃸῕ Ύ᪥Ȃᖹᆒ⃸ 723,; ‒⏝ῄ῟⃺NOMURA400 Ῠ`⃸ญᮧ੷ๆฐࠕᕤᏛ◊"‹ ⁱ‽⁺ῌᥦ౪ῗῦῄ⁅⁺‽ῧῂ ⃸᪥ᮏᰴᘧᕷሙ΅඲ຝ᯶Ή വ ᐃ῕‌῟ᕷሙ௦ࢢᛶ΅዇ῄᢞଘ⁤⁉⁎⁺‷ῧῂ⃺ᵓᡂຝ᯶`᪥ ᮏᰴᘧᕷሙ΅඲ຝ᯶΅୰Ή ⃸†⁈⁨‷⁆΅ព॒‒ᇶΎവᐃ῕‌῟ ᕷሙ௦ࢢᛶ΅዇ῄ 400 ຝ᯶ῧῂ[ญᮧ]⃺῝΅ῆῡ⃸໬Ꮫ⃸๢໠⁹ჰ ๢⃸ᶵᲔ⃸Ҝືత⃸Ⴑᶵ⁹Ɲᐦ⃸་⒪⁹⁖⁩‷ †⃸ᇇရ⃸ᐙᗞ⏝ရ⃸ ၟ♫⃸ᑠ኎ ⃸″⁺⁑‷⃸※⁓⁆․‥†⃸ ⁅‡†⃸೫ಙ⃸೫ಙᘓয⃸ఫ Ꮿ⁹୙ື⏘⃸ഐ౪⃸බ┈⃸ฐࠕ΅ 19 ᴗ✀ῌᑐ૏ῧῂ⃺᪥ᮏ΅ᕷሙ ΅ศᯒΎ`പῗῦῄᣦᶆῠῨุ᩿ῗ῟⃺῾῟⃸ᕷሙ඲య΅ື῍‒ ᢕᥱῙ῟ ΅ᣦᶆῨῗῦ᪥Ȃᖹᆒ⃸TOPIX ண ᣦᶆΎຍῈ῟⃺ ᣦᶆ݅ ΅᫬้‫ ݐ‬ΎῊῑᰴ౯‒‫݅݌‬ǡ ‫ ݐ‬ῨῙῨ⃸༢఩ᮇ࿛᳃‫( ݐ‬1 ⁴᭶⃸2 ⁴᭶⃸3 ⁴᭶)ῧ΅⁨‽⁺ⁱ‫݅ݎ‬ǡ ‫` ݐ‬ୗᘧῧᐃ̿ῧ῍⃺ ‫ݎ‬௜ǡ௧ ൌ 1 ⁴᭶ᚋ΅ண ΎῊῄῦ` 9 ᖺ 9 ⁴᭶΅ 117 ᅇ΅᥎ᐃ‒ࢌΰ῟⃺῾ ῟⃸2 ⁴᭶ᚋ⁹3 ⁴᭶ᚋ΅ண ΅ሙྜ⃸⁨‽⁺ⁱΎῊῑ༢఩ᮇ࿛᳃‫ݐ‬ ‒ਚᩚῙΐῨῧ᥎ᐃ‒ࢌΰ῟⃺ 3.1 1 ⁴᭶ᚋ΅እᤄண Ȇᯝ ូ እᤄண ΅ኀؐṇµ⋡΅Ȇᯝ‒ᅗ 1 Ύ♧Ῑ⃺ΐ΅Ȇᯝ`እᤄண ‒ࢌΰ῟ᅇᩘ΅ῆῡ⃸ኀؐῌ୍Ҥῗῦῄ῟ᅇᩘ΅๭ྜ‒ⓒศ⋡ ῧࢢῗ῟ ΅ῧῂ⃺ 1 ⁴᭶ᚋ΅ண ΅ኀؐṇµ⋡`ᕷሙ඲య΅ືྥ‒ࢢῙ TOPIX ᪥ȂᖹᆒῨῄΰ῟ᕷሙᖹᆒᰴ౯ῧ 63.7%ῧῂΰ῟⃺῾῟⃸ᢞଘุ ᩿Ὸ΅പ⏝ྍЏᛶ΅┠ᏳῨῪ 55%௨ୖ΅ṇµ⋡` 7 ๭௨ୖ΅ᴗ ✀(ண ᣦᶆ 21 ΅ῆῡ⃸15 ᣦᶆ)ῧങᡂῙΐῨῌῧ῍῟⃺55%Ῠῄῆ ኀؐṇµ⋡` AI ‒⏝ῄ῟ᐇၯ΅⁓‟ⁱ⁇ ┠ᣦῗῦῄᩘ್ῧῂ [AERA]⃺ᮏ◊"ῧ`᪥ḟ ࿇ᮇ΅᭶ḟண ῧ⃸ῗΉ ǉ 10 ᖺ࿛Ῠῄῆ࿇ᮇ࿛΅እᤄண ⁄‷⁆ῧᏳᐃῗ῟Ȇᯝ‒♧ῗῦῄ⃺ΐ ‌`ᢞଘᐙΉ ोᮃ΅዇ῄ࿇ᮇண ΅Ɲᗘ‒኱ᖜΎᨵၿῗ῟Ῠῄ ῆΐῨῧῂ⃺ ᅗ 1℆ូ 1 ⁴᭶ᚋ΅እᤄண ΎῊῑኀؐṇµ⋡ ϲϰ͘ϭ ϲϯ͘Ϯ ϲϯ͘Ϯ ϱϵ͘ϴ ϱϵ͘ϴ ϱϵ͘ϴ ϱϵ ϲϬ͘ϳ ϱϳ͘ϯ ϱϳ͘ϯ ϲϬ ϱϱ͘ϲ ϱϲ͘ϰ ϱϲ͘ϰ ϱϳ͘ϯ ኀ ϱϰ͘ϳ ϱϱ͘ϲ ؐ ϱϱ ϱϮ͘ϭ ṇ ϰϵ͘ϲ ϰϴ͘ϳ ϰϳ͘ϵ µ ϱϬ ϲϱ ⋡ ⃲ ‫݌‬௜ǡ௧ା᳃௧ െ ‫݌‬௜ǡ௧ ‫݌‬௜ǡ௧ ϰϱ ϯϱ ഓཤ 3 ᖺ࿛΅᪂Ϊঠ஦Ή ᚓ ‌῟୺ᡂศ‷‱†Ῠᰴ౯⁅⁺‽ ‒⏝ῄῦḟ΅ᅇᖐᘧ‒᥎ᐃῙ⃺ ଵହ ‫ݎ‬௜ǡ௧ ൌ ܽ௜ǡ଴ ൅ ෍ ܽ௜ǡ௞ ‫ݔ‬௞ǡ௧ ௞ୀଵ ូ ‫ݔ‬௞ǡ௧ Ῠ`᫬้‫ݐ‬ΎῊῑ݇୺ᡂศ΅୺ᡂศ‷‱†ῧῂ⃺ᅇᖐศ ᯒ΅ၯ⃸AIC ᇶ‽[Akaike 74]ΎῊῑ‷⁄⁁⁕⁭•‸വᢥ‒ࢌῄ⃸ ਍᫂ຊ΅పῄ਍᫂ኚᩘ`ᅇᖐᘧΎྵ ῦῄῪῄ⃺ ូ 1 ⁴᭶ᚋ΅ண ΎῊῄῦ`⃸ছɾᮇ࿛΅ฌᅇᖐศᯒ΅Ҝ⏤ᗘਚ ᩚ῭῿ỴᐃῨእᤄᮇ࿛΅ኀؐṇµ⋡Ύṇ΅┦࿥ῌ॒ ‌῟⃺῝΅ ᵝᏊ‒ᅗ 2 Ύ♧Ῑ⃺ᅗ୰΅┤ɪ`୍ḟ΅ಶఝ᭤ɪῧῂ⃺ ᅗ 2℆ូ ண ṇµ⋡ῨҜ⏤ᗘਚᩚ῭῿Ỵᐃಀᩘ΅┦࿥ 3. እᤄண Ȇᯝ 2 ÷΅ᡭἲ‒⏝ῄῦእᤄண ΅Ɲᗘ᳨হ‒ࢌΰ῟⃺ᅇᖐᘧ΅᥎ ᐃῊ ῳእᤄண ΅ᡭᅠ`௨ୗ΅೫ ῧῂ⃺ 1998 ᖺ 1 ᭶ 1 ᪥~2000 ᖺ 12 ᭶ 31 ᪥΅ 3 ᖺ࿛(36 ⁴᭶)΅᪂ Ϊঠ஦ῧ਍᫂ኚᩘ΅ছɾ⁅⁺‽‒సᡂῙ⃺ ኀ ؐ ṇ µ ⋡ 1. 2. 3. 4. ྛ᭶΅਍᫂ኚᩘῧ͓᭶ᮎ΅Ǻ್‒ࣅ਍᫂ኚᩘῨῙฌᅇᖐ ᘧ‒᥎ᐃῙ⃺῟ῠῗ⃸ΐ΅᫬᥎ᐃῙ΅` 2.3 ÷ῧ♧ῗ῟⁨ ‽⁺ⁱ΅್ῧῂ⃺ ᥎ᐃ῕‌῟ᅇᖐᘧΎ┤ಶ΅⁄‫⁅⁆‷‫‬⁺‽(2001 ᖺ 1 ᭶ 1 ᪥ ~2001 ᖺ 1 ᭶ 31 ᪥)Ή ᚓ ‌῟୺ᡂศ‷‱†‒௦ධῙΐῨ ῧ͓᭶ᮎ(2001 ᖺ 2 ᭶ᮎ)΅Ǻ್΅᥎ᐃ‒ࢌῆ⃺ ছɾ⁅⁺‽΅సᡂ࿔ጞ᪥‒ 1 ⁴᭶Ὶῤഌ ΊΐῨῧᅇᖐᘧ ‒ẖ᭶᭦᪂ῗῦῄ῍⃸2010 ᖺ 10 ᭶ᮎ῾ῧ΅᥎ᐃ‒ࢌῆ⃺ ϯϵ͘ϯ ϰϬ ໬Ꮫ ๢໠៊ჰ๢ ᶵᲔ Ҝືత Ⴑᶵ៊Ɲᐦ ་۶៊⁖⁩‷ † ᇇရ ᐙᗞ⏝ရ ၟ♫ ᑠ኎ ″⁺⁑‷ ※⁓⁆․‥† ⁅‡† ೫ಙ ᘓয ఫᏯ៊୙ື⏘ ഐ౪ බ┈ ฐࠕ TOPIX ᪥Ȃᖹᆒ ΅༢ਂ‒ྵ ΅`ǉ 10 ୓ǼᏑᅾῗ⃸࿸್௨ୖ΅ඹୣ࿥ಀ΅Ǽ ྜΊ` 400 Ή 500 Ǽῧῂΰ῟⃺ ᜓ ᪥Ȃᖹᆒ ᇇရ ၟ♫ ᘓয බ┈ ་۶࣊ࣝࢫ 723,; ฐࠕ ႱᶵƝᐦ ᶵᲔ ࢯࣇࢺ࢙࢘ ࢧ࣮ࣅࢫ ఫᏯ୙ື⏘ ࣓ࢹ࢕࢔ ๢໠ჰ๢ ഐ౪໬Ꮫ ᐙᗞ⏝ရ Ҝືత ᑠ኎ࡾ ೫ಙ Ҝ⏤ᗘਚᩚ῭ࡳỴᐃಀᩘ Ҝ⏤ᗘਚᩚ῭῿ỴᐃಀᩘῨ`ᅇᖐᘧ΅ᙜῦ`῾ ᗘ‒ࢢῙᣦᶆ ῧῂ ⃸ᅇᖐᘧῌഓཤ΅ኚື‒ ῏਍᫂ῧ῍ῦῄΏῩኀؐṇµ⋡ ῌ዇῏Ὺഴྥῌ॒ ‌῟⃺ΐ‌`ছɾᮇ࿛ῧᙜῦ`῾ ΅ ῄᅇ ᖐᘧ‒సᡂῧ῍ῦῄΏῩእᤄண ΅Ɲᗘ ዇῾ῨῄῆΐῨῧῂ ⃸ -2- The 26th Annual Conference of the Japanese Society for Artificial Intelligence, 2012 ෆᤄ΅ẁၥῧእᤄண ΅Ɲᗘ‒኱῾ΉΎ᥎ῗฎΐῨῌῧ῍⃺῾ ῟⃸ᅗ2Ή ᫂ ΉῪ ῆΎ⃸ᮏ◊"ῧ`ഓ๫പྜῌ⏕ῘῦῄῪῄ⃺ ഓ๫പྜῨ`⃸ছɾᮇ࿛ῧᙜῦ`῾ ΅዇ῄ⁅⁺‽‒సᡂῙῂ῾ Ύỗ⏝ᛶΎḞῑ⃸ᮍ᮶΅⁅⁺‽΅ண Ύ`പ῕ῪῄᙧῨῪΰῦῗ ῾ῆΐῨῧῂ⃺᪂ΪῨῄῆȂ῭ືྥΎ࿥Ῑɉ̦ⓗῪ᝟ሗ‒ᢅῆ⁄ ‫⁅⁆‷‫‬⁺‽Ή ⃸୺ᡂศศᯒΎ ΰῦྜᡂኚᩘ‒సᡂῗ⃸῕ Ύᅇ ᖐศᯒ΅ၯΎ‷⁄⁁⁕⁭•‸വᢥ‒⏝ῄΐῨῧഓ๫പྜῌᅇഺ῕ ‌῟ῨͳῈ ‌⃺ ូ ୍᪉ῧ⃸೫ಙ බ┈ ᘓয⃸་⒪⁹⁖⁩‷ †Ύ࿥ῗῦ`Ҝ⏤ᗘ ਚᩚ῭῿Ỵᐃಀᩘ⃸እᤄண ΅ኀؐṇµ⋡Ῠ ΎపῄỈ‽ῧῂΰ ῟⃺ΐ‌ ΅⏘ᴗῌෆႲ⏘ᴗῨ࿧῰‌⃸᪂Ϊঠ஦Ή ΐ‌ ΅⏘ᴗ Ύ࿥Ῑ≉ᚩฎῌῆ῾῏ᢳฟῧ῍ῪΉΰ῟῟ ῧῂῨͳῈ ‌⃺ ូ ᐇၯΎᢳฟ῕‌῟ඹୣ༢ਂ΅⁘†‒॒ῦ῿Ῠ⃸ᾲ࿇ᮇฐ฼᫢ୖ ᪼ᾳ ᾲ⁅⁓⁪᫢ඞ᭹ᾳῪῩῨῄΰ῟Ȃ῭඲యΎస⏝ῙῨͳῈ ‌ ඹୣ⃸ᾲůᅜ᫢⁆⁦‽Ҝືతᾳ ᾲఫᏯ᫢ࠕଘᾳ΅ ῆΎಶู΅ᴗ✀Ύᙳ ᅚ‒ཬῼῙῨͳῈ ‌ඹୣῌ॒ ‌῟⃺ΐ‌ ΅ฟ⌧⁏‽⁺ⁱ‒ ⏝ῄῦᅇᖐᘧ‒᥎ᐃῗῦῄ῟ ⃸≉ᚩฎῌῆ῾῏ᢳฟῧ῍ᴗ✀ Ῠ῝ῆῧῪῄᴗ✀ῌᏑᅾῙ⃺῝‌ῌᅗ 2 ΅Ҝ⏤ᗘਚᩚ῭῿Ỵᐃ ಀᩘΎࢢ‌ῦῄ⃺ῗΉῗ⃸⁄‫⁆‷‫‬Ῠῄῆே࿛ῌॾจ‒ࢌῆΐῨ΅ῧ ῍᝟ሗ‒⏝ῄῦᰴ౯ண ‒ࢌΰῦῄ῟ ⃸ྛᾫ΅᥎ᐃΎᑐῗῦ Ῡ΅ ῆῪ᝟ሗῌᐇၯΎຠᯝ‒ᣢΰῦῄ῟΅Ή‒ख़ड़ⓗΎ☜Ή ΐῨῌῧ῍⃺ΐ‌`ഓཤ΅ኚືΎᑐῗῦศᯒῌྍЏῧῂ஦⃸ ῝ῗῦᮍ᮶΅ᰴ౯΅᥎ᐃ΅ၯΎ⏝ῄΐῨ΅ῧ῍ྍЏᛶ‒♧ῗῦ ῄ⃺ 3.2 ண ᮇ࿛ู⁹዇ኚືᮇ΅ண ṇµ⋡ 3.1 ÷ῧ` 1 ⁴᭶ᚋ΅እᤄண Ύῤῄῦ೉Ό῟ῌ⃸ᮏᡭἲΎ ΰ ῦῩ΅⛬ᗘඛ΅ᮍ᮶῾ῧ᥎ᐃῧ῍΅Ή‒᳨হῗ῟⃺ᮇ࿛ู΅እ ᤄண ΅ኀؐṇµ⋡‒ᅗ 3 ᕥ΅ᩳɪᲬ‫⁓⁧‮‬ῧ♧Ῑ⃺ᮇ࿛ูῨ` ண ᮇ࿛΅࿇῕ῧῂ ⃸2 ÷ῧ೉Ό῟ ῆΎ 1 ⁴᭶ᚋ⃸2 ⁴᭶ᚋ⃸3 ⁴ ᭶ᚋ΅ 3 ῤ΅ᮇ࿛Ύῤῄῦண Ɲᗘ‒᳨হῗ῟⃺ ូ ண ᮇ࿛ῌ࿇῏ῪΎῤ‌⃸῝΅࿛Ύୣΐ஦૏΅ᩘῌቑῈ῟ ⁄‫⁆‷‫‬ῨȂ῭ኚື΅┦࿥`ᙅ῾⃺ᅇᖐᘧΎ⏝ῄ ‌῟୺ᡂศ ΅॒ῦ῿Ῠ 2 ⁴᭶ᚋ⁹3 ⁴᭶ᚋ΅ண ῧ`⃸1 ⁴᭶ᚋ΅ண ΅ 1.5 ಸΉ 2.0ಸ΅ᩘ΅୺ᡂศῌ⏝ῄ ‌ῦῄ῟⃺ΐ‌`਍᫂ຊ΅ᙅῄ ਍᫂ኚᩘ‒ከ῏⏝ῄΐῨῧȂ῭ኚືΎྜҤῙᅇᖐᘧ‒సᡂῗῦ ῄ῟ΐῨΎῪ ⃸ഓཤ΅ኚື΅਍᫂`ῧ῍ῦ ᮍ᮶΅ኚືΎᑐῙ ண ຊῌؐῡ⃺ᐇၯΎ⃸ᮇ࿛ῌ࿇῏ῪΎῤ‌ῦண Ɲᗘῌᝏ῏ Ὺഴྥῌ॒ ‌῟⃺೛Ύ⃸1 ⁴᭶ᚋ΅ண Ύ࿥ῗῦ 2 ⁴᭶ᚋ⃸3 ⁴ ᭶ᚋ΅ண ᑡῪῄ୺ᡂศΎ ᥎ᐃῧዲᡂʬ‒཰ ῟Ῠῄῆ ΐῨ`⃸῝‌ῠῑȂ῭΅ኚືΎ┤Ȇῗ῟୺ᡂศ‒సᡂῧ῍ῦῄ῟Ῠ ῄῆΐῨῧῂ⃺ ϲϬ ኀ ؐ ṇ µ ⋡ ⃲ ϱϴ ϱϲ ண ᮇ࿛ῌ࿇῏ῪΎῤ‌ῦኀؐṇµ⋡ῌୗῌ୍᪉ῧ⃸ኚື ΅኱῍Ὺ᫬ᮇΎῊῑண ṇµ⋡ῧ` 1 ⁴᭶ᚋ⁹2 ⁴᭶ᚋ΅ண ῧ 2 ⁛•ⁱ⁆௨ୖ΅ᨵၿῌ॒ ‌῟(ᅗ3 ྑ΅ᓞᲬ‫⃺)⁓⁧‮‬ΐ΅ΐῨΉ ᮏᡭἲ` 2 ⁴᭶ᚋ΅ண ῾ῧ᭷ຠῧῂῨῄῆΐῨῌῧ῍⃺ῪῊ⃸ ዇ኚືᮇ΅ண Ῠ`ୗᘧΎ ΰῦᐃ῾⃸᫬้ T Ή ጞ῾ছɾᮇ ࿛ 36 ⁴᭶ΎῊῑ⁨‽⁺ⁱ΅ᶆ‽೫ᕪ ߪ ΅ 0.1 ಸ ȟᑐ್ῌ኱ ῍Ὺኚື‒᥎ᐃῗ῟ሙྜ΅῿ண ‒ࢌΰ῟ ΅ῧῂ⃺ ்ାଷହ ଶ ߪ ଶ ൌ ෍ ሺ‫ݎ‬௜ǡ௧ െ ‫ݎ‬തതതതሻ పǡ௧ ௧ୀ் ΐ΅࿸್Ύ ΰῦᐃ ῟዇ኚືᮇ`඲ண ᮇ࿛΅Ὴ ῝ 3 ศ΅ 1 ῧῂΰ῟⃺ண ῗ῟ᅇᩘ΅ῆῡ⃸ኀؐῌ୍Ҥῗ῟๭ྜ‒ᅗ 3 ΅ᓞῄ Წ‫⁓⁧‮‬ῧ♧ῗῦῄ⃺ 4. ᥦ᱌ᡭἲ΅‵‷⁄⁞໬ ᐇၯ΅ฐࠕ࿥ಀ͵΅౑⏝‒┠ⓗῨῗῦ⃸ᮏ◊"ῧᥦ᱌ῗ῟ᡭἲ ΅‵‷⁄⁞໬‒ࢌΰ῟⃺3 ÷῾ῧ΅ᐇቲῧᢞଘᐙΉ ोᮃ΅዇ῄ࿇ ᮇண ΅᭷ຠᛶ‒☜Ή ῟⃺῝΅ୖῧᢞଘᐙῌ෇⁥Ύᢞଘ‒ࢌῆ ῟ ΅ॾจ‒ࢌῄ⃸ഓཤ΅ኚືΎᑐῗῦ ศᯒῧ῍‵‷⁄⁞‒ᥦ ᱌Ῑ⃺ഓཤ΅ᰴ౯ኚື ⁄‫΅⁆‷‫‬᝟ሗ‒ᥦ♧Ῑ΅῿ῧῪ῏⃸ ศᯒΎᇶῥῄ῟᝟ሗ‒ᥦ౪Ῑ⃺ 4.1 ᥦ᱌‵‷⁄⁞ᴫो ᮏ‵‷⁄⁞` forecast ‽⁔⃸tag cloud ‽⁔⃸ඹୣ࿥ಀฟ⌧⁏‽⁺ ⁱ୍।ῧᵓᡂ῕‌ῦῄ⃺ᅗ 4 ῌ forecast ‽⁔⃸ᅗ 5 ῌ tag cloud ‽ ⁔⃸ᅗ 6 ῌඹୣ࿥ಀฟ⌧⁏‽⁺ⁱ୍।ῧῂ⃺ᮏ÷ῧ`῝‌῞‌ ΅ᅗΎᑐῗῦᶵЏ‒਍᫂Ῑ⃺ ṯforecast ‽⁔ ᣦᶆ῔Ῠ΅ண ΅ኀؐ‒ᰴ౯ኳẼணሗ΅ᙧῧᅗ 4 ᕥ΅ࢢΎ῾Ῠ ῦῄ⃺ᅠΎ 1 ⁴᭶ᚋ⃸2 ⁴᭶ᚋ⃸3 ⁴᭶ᚋ΅ண ‒ࢢ♧ῗ῟ ΅ῧ ῂ ⃸୍┠ῧȂ῭ືྥῌᢕᥱῧ῍ ῆΎῪΰῦῄ⃺ᅗ 4 ྑ΅‫⁧‮‬ ⁓`ᣦᶆ῔ῨΎ‽⁔ῧษ ᭰Ὲ ‌ ῆΎῪΰῦῊ ⃸ᐇၯ΅⁨‽ ⁺ⁱῨᅇᖐῧ᥎ᐃ῕‌῟⁨‽⁺ⁱ‒ছɾᮇ࿛΅ 36 ⁴᭶+ண ΅ 1 ⁴ ᭶ศࢢ♧ῗῦῄ⃺⁤⁺‴`‫ୖ⁓⁧‮‬ൺΎῂ a Ή u ΅‽⁔‒ษ ᭰ῈΐῨῧಶู΅ᣦᶆ‒ศᯒῙΐῨ ྍЏῧῂ⃺῾῟⃸‫⁓⁧‮‬ ୖ΅ྛⅬ‒‭⁨⁁‭ῙῨ⁛⁁⁕†⁁⁕ῌࢢ♧῕‌⃸ྛ᫬ⅬΎῊῄῦ ኚືΎ᭱ ᐤ୚ῗ῟୺ᡂศῌࢢ♧῕‌⃺ኚືΎ᭱ ᐤ୚ῗ῟୺ᡂ ศῨ`⃸῝΅᫬ⅬΎῊῄῦᅇᖐಀᩘῨ୺ᡂศ‷‱†΅ ΅ȟᑐ್ ῌ᭱ ኱῍ῄ୺ᡂศ΅ΐῨῧῂ⃺pre/next ΅⁨ⁱ‭‒‭⁨⁁‭ῙΐῨ ῧ๓᭶/͓᭶΅ศᯒȆᯝΎ⛣ືῙ⃺ ϱϴ͘ϰ ඲ᮇ࿛ ϱϲ͘ϭ ϱϯ͘ϱ ϱϰ ϱϭ͘Ϯ ϱϮ ϱϬ ዇ኚືᮇ ϱϬ͘ϱ ϰϴ͘ϴ ϰϴ ϰϲ ϰϰ ϭ⁴᭶ᚋ Ϯ⁴᭶ᚋ ϯ⁴᭶ᚋ ண ᮇ࿛ ᅗ 3℆ូ ண ᮇ࿛ูண ṇµ⋡⁹዇ኚືᮇ΅ண ṇµ⋡ ᅗ℀℆ូ forecast ‽⁔ -3- The 26th Annual Conference of the Japanese Society for Artificial Intelligence, 2012 ṯtag cloud ‽⁔ ΐ΅᫬ᮇ΅ᅇᖐᘧ΅਍᫂ኚᩘῨῪΰ῟ྛ୺ᡂศΎᑐῗῦ⃸ᅉᏊ૷ ֤ฎ΅኱῍Ὺ༢ਂ‒ྵ ඹୣ༢ਂ⁘†΅୍౛‒୺ᡂศ῔ῨΎ♧ῗ ῦῄ⃺ᅗ 4 ΅‫⁕⁁†⁕⁁⁛΅⁓⁧‮‬ῧࢢ♧῕‌ᙳᅚຊ΅኱῍Ὺ ୺ᡂศῨ⃸ΐ΅ࢢ‒॒ẚΌΐῨῧ⃸Ῡ΅ ῆῪ༢ਂῌῩ΅ᣦᶆ΅Ῡ ΅ኚືΎຠῄῦῄ῟΅ΉਚΌΐῨῌῧ῍⃺῾῟⃸ྛ୺ᡂศ΅୺ ोῪඹୣ༢ਂ‒୍।ῧ῾Ῠ ῦῄ΅`ഓཤ΅ศᯒΎ ᑐᛂῙ ῟ ῧῂ ⃸ΐΐ‒୍॒Ῑ‌῰Ῡ΅ ῆῪȂ῭࿥ೳ΅ঠ஦ῌῂΰ῟ ΅Ή‒኱῾ΉΎᢕᥱῙΐῨῌῧ῍⃺῾῟⃸ྛ୺ᡂศྡ⃴PC1⃸ PC2⃸…⃸PC15⃵`⁨ⁱ‭ᵓ೰‒ᣢΰῦῊ ⃸‭⁨⁁‭ῙΐῨῧᅗ 6 ΅ ඹୣ࿥ಀฟ⌧⁏‽⁺ⁱ‒॒ΐῨῌῧ῍⃺ ᑠ῕ῄ⃺῝΅୍᪉ῧᑡῪῄ୺ᡂศ΅ᩘῧኚື‒਍᫂ῧ῍ῦῄ῟ ⃸ᅗ 5 ᅗ 6 ΅୺ᡂศ୰΅ᅉᏊ૷֤ฎῌ኱῍ῄ༢ਂ‒ྵ ඹୣ ⁘†ῌῩ΅ ῆΎᰴ౯΅ኀؐΎᙳᅚῗῦῄ΅Ή‒ᢕᥱῧ῍⃺ ᮏ‵‷⁄⁞ῧ`ഓཤ΅ᰴ౯⁅⁺‽Ῠ⁄‫⁅⁆‷‫‬⁺‽΅┦࿥ᛶ‒ྍ ख़໬ῙΐῨῧ⃸ॾจ‒ྍЏΎῗῦῄ⃺ഓཤ΅┦࿥Ή ᮍ᮶΅ண ΅ॾจΎᙺWῦΐῨῌᮏ‵‷⁄⁞΅┠ⓗῧῂ⃺ 5. ῾Ῠ ᮏ◊"ῧ` CPR ἲ‒ᛂ⏝ῗῦǉ 10 ᖺ࿛Ῠῄῆ࿇ῄ‷⁏ⁱῧ΅እ ᤄண ‒ࢌΰ῟Ȇᯝ⃸1 ⁴᭶ᚋ΅ண ΎῊῄῦ TOPIX ᪥Ȃᖹᆒ Ῠῄΰ῟ᕷሙᖹᆒᰴ౯ῧ` 60%௨ୖ΅ኀؐṇµ⋡‒཰ ῟⃺῾῟⃸ ᴗ✀ูᖹᆒᰴ౯‒ྵ ῟ኀؐṇµ⋡΅ᖹᆒῧ 55%௨ୖῨῄῆ዇ ῄƝᗘ‒♧ῗ῟⃺ΐ‌`ᐇົῧồ ‌Ỉ‽΅ṇµ⋡ῧῂ[7]⃺ ῕ Ύ⃸ᮏ◊"ῧ⏝ῄ῟ᡭἲ‒ᇶΎ⃸ᢞଘᐙῌॾจ‒ࢌῆΐῨ‒๓ ᥦῨῗ῟Ȃ῭ືྥศᯒ‵‷⁄⁞ ᵓÿῗ῟⃺ΐ΅‵‷⁄⁞Ύ ⃸௒ ῾ῧ୙᫂░ῧῂΰ῟Ȃ῭ኚືῨ⁄‫⁆‷‫‬᝟ሗῨ΅࿥ಀᛶῌ᫂ ΉῨ Ὺΰ῟⃺ᢞଘᐙ`ᮏ‵‷⁄⁞‒⏝ῄῦॾจ‒ࢌῄ⃸ᢞଘάືΎᙺ WῦΐῨῌྍЏῧῂ⃺ ូ ᮏ◊"ῧ⏝ῄ῟ᡭἲΎ`῾ῠᨵၿ΅వᆅῌῂ⃺౛Ὲ῰ᙧែǢ ॾᯒ΅ẁၥῧ ೳਂῌศॾ῕‌ῦῗ῾ΰῦῄⅬ ⃸኱ᐜฎ΅⁅ ⁺‽Ῠῗῦ᪂Ϊঠ஦΅⁅⁺‽ῠῑΎးᐃῗῦῗ῾ΰῦῄⅬῧῂ⃺ ᐇၯ΅ᢞଘΎ`ᵝᾫῪ᝟ሗ‒⏝ῄῦॾจ‒ࢌῆ῟ ⃸༢୍΅᝟ሗ ῧ`Ὺ῏ࣽᩘᣦᶆΎᑐᛂῗ῟‵‷⁄⁞‒ᥦ᱌ῗῦῄ῍῟ῄ⃺ ཧͳᩥ⊩ ᅗ 5℆ូ tag cloud ‽⁔ ṯඹୣ࿥ಀฟ⌧⁏‽⁺ⁱ୍। ছɾᮇ࿛⁹እᤄᮇ࿛ΎῊῑ୺ो༢ਂ⁘†΅᫬ǄิⓗῪฟ⌧⁏‽ ⁺ⁱ᝟ሗ‒♧ῗῦῄ⃺୺ो༢ਂῨ`ྛ୺ᡂศΎῊῄῦᅉᏊ૷֤ ฎῌୖ఩ 15 ఩῾ῧ΅ ΅Ῠῗ῟⃺῝΅୺ो༢ਂ‒ྵ ඹୣ༢ਂ⁘ †඲ῦ΅ฟ⌧⁏‽⁺ⁱ‒ᮏ‵‷⁄⁞ῧ`ࢢ♧ῗ⃸ୖΎῂΏῩᅉᏊ ૷֤ฎῌ኱῍ῄ༢ਂ‒ྵ ⁘†ῨῪΰῦῄ⃺ΐ΅ฟ⌧΅᭷↓Ῠᰴ ౯΅ኀؐ‒ྜ‫‏‬ΊῦศᯒῙΐῨῧ⃸Ῡ΅ ῆῪ⁘†΅ฟ⌧΅᭷↓ ῧኀؐῌỴ῾ΰῦῄ΅Ή‒ᢕᥱῙΐῨῌῧ῍⃺ [AERA] AERA’2012.2.13⃸pp.62⃸ᮅ᪥᪂Ϊฟ∧. [Akaike 74] Akaike, H.: A new lool at the statistical model identification, IEEE Transactions on Automatic Control, Vol.19, pp.716-723 (1974). [ChaS] ChaSen ⁙⁺⁞⁘⁺‶ : http://chasen.naist.jp/hiki/chasen . [዇ 2002] ዇ὒ⃸֘஭ၟࢌ⃸኱yᩒ⃸⏣୰࢘ ℆ ⁉⁣⁺⁧⁩⁋ ⁁⁆⁭⁺‭Ύ ᫬ᮇ΅ᰴ౯ண ⃹ᰴ౯ண ΎῊῑ⁓‡⁩‽⁨ⁱ‫‮‬ Ύ ≉ᚩฎᢳฟ⃹⃸ႱᏊ᝟ሗ೫ಙᏛ఍ᢏ࢑◊"ሗ(NPL)⃸ Vol.102⃸No.432⃸ pp.13-16 (2002) . [ఀᗞ 2006] ఀᗞᩧᚿ ℆ ೽໬ਭⓗᡭἲ‒⏝ῄ῟ฐࠕ⁅⁺‽΅ண ⃸᪥ᮏಙᆇᛶᏛ఍৵⃸Vol.28⃸No.7⃸pp.471-480 (2006) . [࿴Ἠ 2011] ࿴Ἠ₩⃸ᚋ‫ܘ‬༟⃸ᯇ஭‫ܘ‬஬൫℆ Ȃ῭⁄‫⁆‷‫‬᝟ሗ‒⏝ ῄ῟࿇ᮇⓗῪᕷሙືྥ᥎ᐃ⃸᝟ሗฎ⌮Ꮫ఍৵⃸Vol.52⃸No.12⃸ pp.3309-3315 (2011) . [኱⃝ 2006] ኱⃝ᖾ⏕ : ‿⁡ⁱ‷Ⓨ॒΅⁅⁺‽ศᯒ᫢⁠⁅⁩໬+ ྍख़໬+‱⁝⁣⁉ ⁺‵⁥ⁱ᮫‵⁈⁨ ๰Ⓨ⃸ᮾிႱᶵ኱Ꮫฟ∧ᒁ (2006). ᅗ 6℆ូ ඹୣ࿥ಀฟ⌧⁏‽⁺ⁱ୍। 4.2 ᮏ‵‷⁄⁞‒⏝ῄ῟Ȃ῭ືྥศᯒ΅฼⏝౛ ᭱ึΎ forecast‽⁔΅ᬕ/Ⴅ΅ࢢ୍।ῧ௒ᚋ΅Ȃ῭⁆⁪ⁱ⁇‒୍┠ ῧᢕᥱῧ῍ ῆΎῪΰῦῄ⃺‫⁓⁧‮‬Ύῤῄῦ`ഓཤ΅ኚືΎᑐ ῗῦෆᤄ࿥ᩘ‒సΰῦῄῌ⃸Ҝ⏤ᗘਚᩚ῭῿Ỵᐃಀᩘ`ẚుⓗ [ ᪥Ȃ] ᪥Ȃ‵ ※ ⁺ ⁧ ‷ ⃸ ᪥ᮏȂ῭᪂Ϊ⁅ ‶ ‽ ⁩ ⁅ ‡ † http://vip-test2.nikkei.co.jp/help/contract/price/02/help_KIJI_thes.html [ญᮧ] ญᮧ੷ๆฐࠕᕤᏛ◊"‹ⁱ‽⁺⁙⁺⁞⁘⁺‶ : http:// qr.nomura.co.jp/jp/n40/index.htmlᬌᬌ ᬌᬌ῟ῠῗ⃸12085$ ΅ᥦ౪` ᖺ ᭶ᮎ᪥‒ ΰῦ Ǻ஢ῗῦῄ῾Ῑ⃺ូ -4-