ேᩥ⛉Ꮫ⛉ ᩍ⫱◊✲ᨭࢭࣥࢱ࣮
ሗ⛉Ꮫⓗࣉ࣮ࣟࢳࡼࡿ
ⱥㄒ᳨ᐃᩍ⛉᭩ࡢㄒᙡศᯒ
ඵ㫽 ྜྷ᫂* 㔝 ె௦Ꮚ**
㸦2016ᖺ11᭶24᪥ཷ⌮㸧
㸯㸬ࡣࡌࡵ
♫ࡢࢢ࣮ࣟࣂࣝࡀᣦࡉࢀࡿ୰ࠊᏛ⏕ࡸ◊✲⪅ࠊ ᢏ⾡⪅࡞ࡣࠊࡑࢀࡒࢀࡢάື┠ⓗᑐᛂࡍࡿࡓࡵࡢ ⱥㄒຊࡢྥୖࡀồࡵࡽࢀࡿᶵࡀከࡃ࡞ࡗ࡚࠸ࡿࠋ⥲ྜ ⓗ࡞ⱥㄒຊࡢ⫱ᡂࡢࡓࡵࡣࠊ」ྜⓗ࡞ⱥㄒᏛ⩦ࡀᚲせ ࡞ࡿࡀࠊࡑࢀࡽࢆᨭ࠼ࡿ㔜せ࡞ᰕࡢࡘࡣㄒᙡࡢᏛ⩦ ࡛࠶ࡿゝࡗ࡚ࡼ࠸ࡔࢁ࠺ࠋ ⚾ࡓࡕࡣࠊ༢ㄒࡣᩥ⬦ࡢ୰࡛ぬ࠼ࡿࡇࡀຠᯝⓗ࡛ ࠶ࡿ࠸࠺➨ゝㄒ⩦ᚓㄽࡢ▱ぢ㸦ⓑ㸧ࢆ㋃ ࡲ࠼ࠊⱥㄒࡢᤵᴗࢆ㏻ࡋࡓㄒᙡ⩦ᚓࡢຠᯝⓗ᪉ἲࡘ࠸ ࡚◊✲ࢆ⾜ࡗ࡚࠸ࡿࠋࡲࡓࠊㄒᙡࡢ◊✲࡛ࡣつᶍ࡞ ࢹ࣮ࢱࡢゎᯒࡀᚲせ࡞ࡿሙ㠃ࡶከ࠸ࡓࡵࠊ◊✲᪉ἲ ࡋ࡚ࡣࠊⱥㄒᩍ⫱ሗฎ⌮࡞ࡢሗ⛉Ꮫⓗࣉ࣮ࣟ ࢳ ࢆ ✚ᴟ ⓗ ྲྀࡾ ධ ࢀ࡚ ࠸ࡿ 㸦ඵ 㫽 ࣭㔝 ࠕ㧗 ᑓ ⏕ࠖ㸧ࠋ ㏆ᖺࠊⱥㄒࡢㄒᙡ◊✲࡛ࡣࠊ⮬↛ゝㄒฎ⌮ࡢᙧែ⣲ ゎᯒࢶ࣮ࣝࢆά⏝ࡋࡓሗ࿌ࡀቑຍࡋ࡚࠸ࡿ㸦⏣୰࣭ ᾆ࣭ᚨぢ㸹ᒾᓮ㸧ࠋࡶࡶⱥᩥࡣࠊ༢ㄒ༢ㄒࡢ㛫 ༊ษࡾᩥᏐࡋ࡚ࡢ✵ⓑࢆධࢀ࡚ᩥ❶ࢆసᡂࡍࡿࡓࡵࠊ ࢥࣥࣆ࣮ࣗࢱୖ࡛ฎ⌮ࡋࡸࡍ࠸ࢹ࣮ࢱ࡞ࡗ࡚࠸ࡿࠋᙧ ែ⣲ゎᯒࢶ࣮ࣝࡣࠊⱥᩥࡢᩥ⬦ࡽ༢ㄒࡢရモࢆ⮬ື࡛ ุ᩿ࡋࠊㄒᑿኚࡋࡓ༢ㄒࢆᇶᮏᙧᡠࡍ࠸࠺࣐ࣞ 㸦lemmatization㸧ࡢᶵ⬟ࢆᣢࡗ࡚࠸ࡿ㸦⏣୰㸧ࠋ ࡇ࠺ࡋࡓᢏ⾡ⓗ⫼ᬒࢆ㋃ࡲ࠼ࠊᅇࡢ◊✲࡛ࡣࠊᤵ ᴗᩍᮦࡢⱥᩥࢆศᯒࡍࡿᡭẁࡋ࡚ࠊᙧែ⣲ゎᯒࢶ࣮ࣝ ࡢά⏝ࡑࡢ᭷ຠᛶࡢ᳨ドࢆヨࡳ࡚ࡳࡿࡇࡋࡓࠋ㸰㸬◊✲┠ⓗ
ᮏ◊✲ࡣࠊᤵᴗᩍᮦࡢⱥᩥࢆᵓᡂࡍࡿ༢ㄒࢆࠊሗ ฎ⌮ᢏ⾡ࢆ⏝࠸࡚ຠ⋡ⓗศᯒࡍࡿ᪉ἲࢆぢฟࡍࡇࢆ ┠ⓗࡋ࡚࠸ࡿࠋ ⚾ࡓࡕࡣࡇࢀࡲ࡛ࠊᩥ⬦ᇶ࡙࠸ࡓ༢ㄒᏛ⩦ࢆಁ㐍 ࡍࡿࡓࡵࠊᤵᴗᩍᮦࡽ㔜せ༢ㄒࢆ㑅ᢥࡋࠊᏛ⏕ᥦ ♧ࡍࡿ ᪉ἲࡘ࠸࡚◊✲ࢆ⾜ࡗ࡚ࡁࡓ 㸦ඵ㫽࣭㔝 ࠕⱥ༢ㄒᏛ⩦⏝ࠖ㸧ࠋⱥᩥࢆᵓᡂࡍࡿ༢ㄒࢆ㑅ูࡍࡿ㝿 ࡣࠊ๓ฎ⌮ࡋ࡚ࠊྛ༢ㄒࢆᇶᮏᙧᡠࡍᚲせࡀ࠶ ࡿࡀࠊᚑ᮶ࠊࡑࡢసᴗࡣேᡭࡼࡾ⾜ࡗ࡚ࡁࡓࠋᅇࠊ ࡇࡢసᴗࢆ⮬ືࡍࡿࡓࡵࠊ⮬↛ゝㄒฎ⌮ᢏ⾡ࢆヨ⏝ ࡍࡿࠋࡲࡓࠊ⮬ື࡛ᇶᮏᙧᡠࡋࡓ༢ㄒࡢ᳨ドసᴗࡣࠊ ඛ⾜◊✲࡛㛤Ⓨ῭ࡳࡢⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ࣒㸦ᪧ ⛠㸸ⱥ༢ㄒ㑅ᢥࢩࢫࢸ࣒㸧 ࢆ⏝ࡍࡿ 㸦ඵ㫽࣭㔝 ࠕⱥ༢ㄒ㑅ᢥࠖ㸧ࠋࡇࡢࢩࢫࢸ࣒ࡣࠊⶶࡍࡿࢹ࣮ࢱ ࣮࣋ࢫࡼࡾࠊㄪᰝᑐ㇟ࡢ༢ㄒ㛵ࡋ࡚ࠊ」ᩘࡢⱥ༢ㄒ ᮏ࠾ࡅࡿᥖ㍕≧ἣࢆ⮬ື࡛⾲♧ࡍࡿᶵ⬟ࢆ᭷ࡋ࡚࠸ࡿࠋ ࡇࢀࡽࡢሗ⛉Ꮫⓗࣉ࣮ࣟࢳࢆά⏝ࡍࡿࡇ࡛ࠊ ⱥᩥࡢㄒᙡศᯒࢆຠ⋡ⓗ⾜࠺᪉ἲࢆ☜❧ࡍࡿࡇࡀࠊ ᮏ◊✲ࡢ≺࠸࡛࠶ࡿࠋ㸱㸬ᐇ㦂᪉ἲ
ᅇࡢㄪᰝ࡛ࡣࠊ⩌㤿ᕤᴗ㧗➼ᑓ㛛Ꮫᰯ㸦௨ୗࠊ⩌ 㤿㧗ᑓ㸧ࡢᤵᴗ࡛ᐇ㝿⏝ࡉࢀ࡚࠸ࡿⱥᩥᩍᮦࢆᐇ㦂 ࢹ࣮ࢱࡋࠊᙧែ⣲ゎᯒࢶ࣮ࣝࢆࡗ࡚⮬ື࡛ᇶᮏᙧ ᡠࡋࡓ༢ㄒࡀࠊⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ࣒㐺⏝ྍ⬟ ࠺ࢆホ౯ࡍࡿࠋࡑࡢᐇ㦂᪉ἲࢆ௨ୗ♧ࡍࠋ 㸱㸬㸯 ᐇ㦂ࢹ࣮ࢱ ᐇ 㦂 ࢹ ࣮ ࢱ ࡣ ᳨ ᐃ ᩍ ⛉ ᭩ ࡢ ࠗ MY WAY English Communication I ࠘㸦᳃ఫࠊ㸧࡛࠶ࡿࠋᖺᗘࡢ⩌ 㤿㧗ᑓᖺḟࡢࠕⱥㄒ㸿ࠖࡢᣦᐃᩍᮦ࡛࠶ࡾࠊࡇࡢᩍ⛉ ᭩ࡢᮏᩥࢹ࣮ࢱࢆ⏝ࡍࡿࠋ ᅇࡢฎ⌮⏝ࡍࡿࢹ࣮ࢱ㔞ࡣࠊᤵᴗᅇศࡢ⠊ᅖࡢ ⱥᩥࡢ༢ㄒᩘࡋࠊᤵᴗᅇᩘᅇศ㸦ᖹᡂᖺ᭶᪥ ࡽ᭶᪥ࡲ࡛㸧ࡢࢹ࣮ࢱࢆ⏝ࡍࡿࠋ 㸱㸬㸰 ᙧែ⣲ゎᯒࢶ࣮ࣝࡼࡿ༢ㄒࡢ࣐ࣞ ᩍ⛉᭩ࡢⱥᩥᑐࡋ࡚ࠊ⮬↛ゝㄒฎ⌮ࡢᙧែ⣲ゎᯒ ࢶ࣮࡛ࣝ࠶ࡿTreeTagger㸦Schmid“Probabilistic”㸧ࢆᐇ⾜ࡋࠊ༢ㄒࢆᇶᮏᙧᡠࡋࡓࣜࢫࢺࢆసᡂࡍࡿࠋࡇࡢ ࢶ࣮ࣝࡣࠊࢻࢶࡢᩘ⌮ゝㄒᏛࡢ◊✲⪅࡛࠶ࡿHelmut Schmid ࡀ 㛤 Ⓨ ࡋ ࠊ බ 㛤 ࡋ ࡚ ࠸ ࡿ ࡶ ࡢ ࡛ ࠶ ࡿ 㸦 ⏣ ୰ 㸧ࠋWindows OS ࡸ PC-Linux OSࠊMac OS࡞࡛ᐇ ⾜ ྍ ⬟ ࡛ ࠶ ࡿ ࡀ 㸦 Schmid “TreeTagger” 㸧 ࠊ ᅇ ࡣ Windows OSୖ࡛ᐇ⾜⎔ቃࢆᩚ࠼ࡓࠋ㧗㏿ฎ⌮ࡀ⾜࠼ ࡿࢶ࣮࡛ࣝ࠶ࡾࠊㄒࡢⱥᩥࢆ⣙⛊࡛ゎᯒ࡛ࡁࡿ 㸦OS: Windows7,CPU: 2.3GHz, RAM: 4GB㸧ࠋ
TreeTaggerࡢᐇ⾜ᡭ㡰ࢆᅗ♧ࡍࠋධຊࢹ࣮ࢱࡣࠊ ࢸ࢟ࢫࢺࣇࣝᙧᘧࡢⱥᩥ࡛࠶ࡿࠋࣃࢯࢥࣥୖ࡛ࢥ ࣐ࣥࢻࣉࣟࣥࣉࢺࢆ㉳ືࡋࠊࢫࢡࣜࣉࢺࣇ࡛ࣝ࠶ࡿ “tag-english”ࢆᐇ⾜ࡍࡿࠋᐇ⾜ࡢ➨ᘬᩘࡣࠊⱥᩥࡢࢸ ࢟ࢫࢺࣇࣝྡ࡛࠶ࡿࠋᐇ⾜⤖ᯝࡣࠊࢹࣇ࢛ࣝࢺ࡛ࡣ ᐇ⾜⏬㠃ୖᶆ‽ฟຊࡉࢀࡿタᐃ࡞ࡗ࡚࠸ࡿࡓࡵࠊࣜ ࢲࣞࢡࢺᶵ⬟ࢆࡗ࡚ࠊࢸ࢟ࢫࢺࣇࣝฟຊࡉࡏ ࡿࠋฟຊࣇࣝࡣࠊ༢ㄒࡢゎᯒ⤖ᯝࡋ࡚ࠊ⾜ ༢ㄒရモ␎ྕࠊ༢ㄒࡢᇶᮏᙧࡀฟຊࡉࢀࡿࠋရモ␎ྕ ࡋ࡚ࡣࠊྡモࡢ༢ᩘᙧࡲࡓࡣ㉁㔞ྡモࡣNN㸦noun, singular or mass 㸧 ࠊ ྡ モ ࡢ 」 ᩘ ᙧ ࡣ NNS 㸦 noun, plural㸧ࠊືモࡢཎᙧࡣVV㸦verb, base form㸧ࠊືモࡢ ືྡモ㸭⌧ᅾศモࡣVVG㸦verb, gerund/participle㸧ࠊ ືモࡢ㐣ཤศモࡣVVN㸦verb,past participle㸧࡞ࡀ ࢃࢀ࡚࠸ࡿ㸦UNIVERSITY㸧ࠋ 㸱㸬㸱 ⱥᩥศࣉࣟࢢ࣒ࣛࡢ㛤Ⓨ TreeTaggerࡢฟຊࣇࣝࡣࠊⱥᩥฟ⌧ࡍࡿ࡚ ࡢ༢ㄒࡸྃㄞⅬࠊグྕࠊࣛࣅᩘᏐ࡞ࡢゎᯒ⤖ᯝࡀ ࡑࢀࡒࢀ⾜ࡎࡘฟຊࡉࢀࡿࡓࡵࠊࡑࡢࡲࡲࡢ≧ែ࡛ ࡣᩍ⛉᭩ࡢⱥᩥ༢ㄒࡢᑐᛂࡅࡀᐜ࡛᫆ࡣ࡞࠸࠸࠺ ၥ㢟ࡀ࠶ࡿࠋࡇࡢၥ㢟ࢆゎỴࡍࡿࡓࡵࠊⱥᩥࢆᵓᡂࡍ ࡿ༢ㄒࢆศࡍࡿࣉࣟࢢ࣒ࣛࢆࠊExcelࡢVisual Basic for Applications (VBA)ࢆࡗ࡚㛤Ⓨࡋ࡚ࠊࡑࡢᑐᛂࡅస ᴗࢆ⿵ຓࡍࡿࣜࢫࢺࢆ⮬ື࡛సᡂࡍࡿࡇࡋࡓࠋ 㛤Ⓨࣉࣟࢢ࣒ࣛࡢᐇ⾜⏬㠃ࢆᅗ♧ࡍࠋࡇࡢࣉࣟ ࢢ࣒ࣛࡣࠊࢹ࣮ࢱࡢᩚᙧࢆࡋࡓࢡࣞࣥࢪࣥࢢసᴗࡸࠊ ࣉࣟࢢ࣒ࣛࡀฟຊࡍࡿ༢ㄒᩘࡢ㞟ィసᴗࢆ⮬ືࡍࡿࡶ ࡢ࡛࠶ࡿࠋᤵᴗᅇศࡢⱥᩥࢆධຊࡍࡿࠊⱥᩥࡢࣃࣛ ࢢࣛࣇࢭࣥࢸࣥࢫࡈࢼࣥࣂࣜࣥࢢࢆ⾜࠺ࠋࡑࡢᚋࠊ ᩥ❶ࢆࢭࣥࢸࣥࢫࡈศࡅ࡚ࠊࡑࢀࡒࢀࡢᩥ❶ࡢ┤ୗ ࠊฟ⌧ࡋࡓ༢ㄒࢆ୪࡚ࣜࢫࢺࡋ࡚ฟຊࡍࡿࠋࡇࡢ ࠊྃㄞⅬࡸグྕࠊࣛࣅᩘᏐ࡞ࡣฟຊࡢᑐ㇟እ ࡍࡿࠋࡲࡓࠊᅇࡢᤵᴗ⠊ᅖ࡛ྠ୍ࡢ༢ㄒࢆ⧞ࡾ㏉ࡋ㔜 せ༢ㄒࡋ࡚㑅ࡧฟࡍࡇࡣ࡞࠸ࡓࡵࠊ㔜」ࡋ࡚ฟ⌧ࡍ ࡿ༢ㄒࢆࣜࢫࢺࡽ๐㝖ࡍࡿᶵ⬟ࢆᣢࡓࡏࡿࠋ ࡇࡢࣉࣟࢢ࣒ࣛࢆ⏝ࡍࡿࡇ࡛ࠊศࡉࢀࡓಶࠎ ࡢ༢ㄒࡀᩍ⛉᭩ࡢࡢᩥ❶ࡽ㑅ࡤࢀࡓࡶࡢ࡛࠶ࡿࢆ ♧ࡍࠊ࢞ࢻⓗᙺࢆᯝࡓࡍࣜࢫࢺࢆసᡂࡍࡿࡇࡀ࡛ ࡁࡿࠋ ḟࠊ㛤Ⓨࡋࡓⱥᩥศࣉࣟࢢ࣒ࣛࡢᐇ⾜⤖ᯝࡢ ࢆᅗ♧ࡍࠋࣜࢫࢺࡢᕥഃࡣࠊฟ⌧ࡋࡓ࡚ࡢ༢ ㄒࢆ⾲♧ࡍࡿࠋࣜࢫࢺࡢྑഃࡣࠊ㔜」ࡍࡿ༢ㄒࢆ๐㝖 ࡋࡓࣜࢫࢺࠊྛ༢ㄒࡢฟ⌧ᅇᩘࢆ⾲♧ࡍࡿࠋฟ⌧㢖ᗘ ࡢ್ࢆ⮬ື࡛㞟ィࡋ࡚⾲♧ࡍࡿࡇ࡛ࠊ⏝⪅ࡀ㔜せ༢ ㄒࢆ㑅ᢥࡍࡿ㝿ࡢุ᩿ᮦᩱࡢࡘࡍࡿࡇࡀ࡛ࡁࡿࠋ ᅗ ⱥᩥศࣉࣟࢢ࣒ࣛᐇ⾜⏬㠃 ো ৡ 崯 嵤 崧 ؟ ઇ ఐ છ 峘 ஶ ધ ৰ ষ એ ؟ 崛 嵆 嵛 崱 峼 ো ৡ ৰ ষ ટ ؟ ল ৡ 崯 嵤 崧 ੪峘ୁ岝ષဨ岝লৡୁ੦মOHPPD WDJHQJOLVK LQW[W!RXWW[W
ᅗ TreeTagger ᐇ⾜ᡭ㡰 /HVVRQ:ULWLQJ6\VWHPVLQWKH:RUOG /HVVRQ:ULWLQJ6\VWHPVLQWKH:RUOG >@ ⋇'R\RXUHDGRUZULWHVRPHWKLQJHYHU\GD\">@⋇'R\RXUHDGRUZULWHVRPHWKLQJHYHU\GD\" 'R 'R \RX \RX UHDG UHDG RU RU ZULWH ZULWH VRPHWKLQJ VRPHWKLQJ HYHU\ HYHU\ GD\ GD\ ⋈<RXSUREDEO\HQMR\UHDGLQJERRNV ⋈<RXSUREDEO\HQMR\UHDGLQJERRNV <RX SUREDEO\ SUREDEO\ HQMR\ HQMR\ UHDGLQJ UHDGLQJ ERRNV ERRNV ⋉<RXPD\VRPHWLPHVZULWHHPDLOVWR\RXUIULHQGV ⋉<RXPD\VRPHWLPHVZULWHHPDLOVWR\RXU IULHQGV PD\ <RX VRPHWLPHV PD\ HPDLOV VRPHWLPHV WR ZULWH \RXU HPDLOV IULHQGV WR >@⋇8VLQJOHWWHUVDQGFKDUDFWHUVLVLPSRUWDQWIRUFRPPXQLFDWLRQ \RXU 8VLQJ IULHQGV OHWWHUV >@ ⋇8VLQJOHWWHUVDQGFKDUDFWHUVLVLPSRUWDQW IRU FRPPXQLFDWLRQ DQG ؟ളୁ ځୁ峘লਠਯ 㔜」༢ㄒ๐㝖๓ 㔜」༢ㄒ๐㝖ᚋ ᅗ ⱥᩥศࣉࣟࢢ࣒ࣛᐇ⾜⤖ᯝ ࠉࠉ7+(*810$̻.2+6(15(9,(:1R
㸱㸬㸲 ㄒᙡศᯒ
ᐇ㦂ࢹ࣮ࢱ࡛࠶ࡿⱥᩥࢆTreeTagger㐺⏝ࡋ࡚࣐ࣞ ࡉࢀࡓ༢ㄒࠊⱥᩥศࣉࣟࢢ࣒ࣛࢆࡋ࡚ฟຊࡉࢀࡓ ⱥᩥࡁ༢ㄒࣜࢫࢺࡢࢹ࣮ࢱࢆ࣐࣮ࢪࡍࡿసᴗࢆேᡭ ࡼࡾ⾜࠸ࠊⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ࣒ࡢධຊࢹ࣮ࢱ࡞ ࡿ༢ㄒࣜࢫࢺࢆసᡂࡍࡿࠋ ࡇࡢࣜࢫࢺࡢ༢ㄒ㒊ศࢆධຊࢹ࣮ࢱࡋ࡚ࠊⱥ༢ㄒ ศᯒ⾲సᡂࢩࢫࢸ࣒ࢆᐇ⾜ࡍࡿࠋࡑࡢ⤖ᯝࠊㄪᰝᑐ㇟ࡢ ༢ㄒࡑࢀࡒࢀᑐࡋ࡚ࠊ」ᩘࡢⱥ༢ㄒᮏࡢ࠺ࡕぢฟࡋㄒ ᢅ࠸࡛ᥖ㍕ࡉࢀ࡚࠸ࡿᩘࢆ⮬ື㞟ィࡋࡓ್ࡀࠊ୍ぴ⾲ ࡋ࡚ฟຊࡉࢀࡿࠋࡉࡽࠊࡑࢀࡒࢀࡢ༢ㄒࡢヲ⣽ሗ ࡋ࡚ࠊྛ༢ㄒᮏᥖ㍕ࡉࢀ࡚࠸ࡿ✀㢮ࡀࠊぢฟࡋㄒᢅ ࠸ࡢሙྜࡣ༳ࠊὴ⏕ㄒ࣭㛵㐃ㄒࡢሙྜࡣ୕ゅ༳࡛ ⾲♧ࡉࢀࡿ㸦ඵ㫽࣭㔝ࠕⱥ༢ㄒ㑅ᢥࠖ㸧ࠋ ࡇ࠺ࡋ࡚సᡂࡉࢀࡓ༢ㄒࡢศᯒ⤖ᯝࢆ♧ࡍ⾲㸦༢ㄒ ศᯒ⾲㸧ࢆࡶࠊⱥᩥࡢᩥ⬦ࢆ☜ㄆࡋ࡞ࡀࡽࠊ㔜せ༢ ㄒࢆ㑅ᢥࡋ࡚࠸ࡃࠋ ᐇ⾜ࡋ࡚ࠊ༢ㄒศᯒ⾲㑅ู⤖ᯝࡢ୍㒊ࢆ⾲ ♧ࡍࠋࣜࢫࢺࡢᕥഃ༢ㄒࡢࠕ㑅ู⤖ᯝࠖࢆグධࡍࡿ ࡓࡵࡢ༊ศḍࢆタࡅࠊ㔜せ༢ㄒࡋ࡚㑅ᢥࡋࡓ༢ㄒ 㔜༳ࢆࡅࡿࠋࡑࡢࠊ୰Ꮫ᪤⩦ㄒࡣᫍ༳ࢆࠊ㔜せ ༢ㄒࢆ㝖ࡃᩍ⛉᭩᪂ฟㄒ࡞ࡣ୕ゅ༳ࢆࡅ࡚࠸ࡃࠋ㸲㸬ᐇ㦂⤖ᯝ⪃ᐹ
௨ୖ㏙ࡓㄒᙡศᯒసᴗࡢྛ㐣⛬ࢆࠊ᳨ᐃᩍ⛉᭩ ࡢⱥᩥࢆ⏝࠸࡚ᐇ㝿᳨ドࡍࡿࡇ࡛ࠊTreeTaggerࡼ ࡾ⮬ື࡛࣐ࣞࡋࡓⱥ༢ㄒࡢࠊⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ ࣒ࡢ㐺⏝ྍ⬟ᛶࢆホ౯ࡋࡓࠋ௨ୗࠊࡑࡢ⤖ᯝࢆ㏙ ࡿࠋ ᩍ⛉᭩ࡢᮏᩥࢆࠊᤵᴗᅇศ┦ᙜࡍࡿ⣙ㄒࡈ ศࡋࠊྜィᅇศࡢⱥᩥࢆࡗ࡚ᐇ㦂ࢆ⾜ࡗࡓࠋᐇ㦂 ⏝ࡋࡓ༢ㄒࡢ⥲ᩘࡣࠊㄒ࡛࠶ࡿࠋࡇࢀࡽࡢ༢ ㄒࢆࠊTreeTaggerࢆ⏝ࡋ࡚⮬ື࡛ᇶᮏᙧᡠࡋࡓᚋࠊ ᐇ㦂ᡭ㡰ᚑࡗ࡚㉁ⓗศᯒࡋࡓ⤖ᯝࠊ㑅ࡧฟࡋࡓ㔜せ ༢ㄒࡣㄒ㸦࠺ࡕࠊ୰Ꮫ᪤⩦ㄒࡣㄒ㸧࡞ࡗࡓࠋ ࡲࡓࠊᅇࡢᐇ㦂ࢆ㏻ࡋ࡚ࠊTreeTagger࡛࣐ࣞࡋࡓ ༢ㄒࢆⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ࣒ࡢධຊࢹ࣮ࢱࡋ࡚ ⏝ࡍࡿࡓࡵࡣࠊḟ♧ࡍ๓ฎ⌮ࢆ⾜࠺ᚲせࡀ࠶ࡿࡇ ࡀุ᫂ࡋࡓࠋ ࢱࢺࣝ⾜ࡢ༢ㄒࡢඛ㢌ᑠᩥᏐ ᅇࡢᐇ㦂ࢹ࣮ࢱ࡛ࡣࠊࢱࢺࣝ⾜⏝ࡉࢀࡿྛ ༢ㄒࡣࠊ୍㒊ࡢෙモࡸ๓⨨モࠊ᥋⥆モ࡞ࢆ㝖ࡁࠊඛ㢌 ᩥᏐࡀᩥᏐ࡛᭩ࢀ࡚࠸ࡿࡓࡵࠊᅛ᭷ྡモぢ࡞ࡉࢀ ࡚ࡋࡲ࠸ࠊ༢ㄒࡀ」ᩘᙧࡢሙྜ࡛ࡶࠊTreeTagger࡛࣐ࣞ ࡉࢀࡿ㝿ㄒᑿࡢ“s”ࡀ๐㝖ࡉࢀ࡞࠸ࠋࡑࡢࡲࡲࡢ≧ ែ࡛ⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ࣒ࢆᐇ⾜ࡍࡿࠊᥖ㍕ᩘࡀ ࢮࣟ࡞ࡿ☜⋡ࡀ㧗࠸ࠋࡑࡢࢆ⾲♧ࡍࠋࢩࢫࢸ ࣒ࡢཧ↷ࢹ࣮ࢱ࣮࣋ࢫ࡛࠶ࡿⱥ༢ㄒᮏ࡛ࡣࠊࢇࡢ ༢ ㄒࡀ ぢฟࡋㄒࡋ࡚ࡣ༢ᩘᙧ࡛ᥖ㍕ࡉࢀ࡚ ࠾ ࡾ ࠊ “clothes”ࡸ“means”࡞ࡢ୍㒊ࡢእࢆ㝖ࡁࠊㄒᑿ“s” ࡀࡃ」ᩘᙧࡣᥖ㍕ࡉࢀ࡚࠸࡞࠸ࠋ ࡇࡢၥ㢟ࡢᑐ⟇ࡋ࡚ࠊࢱࢺࣝ⾜ࡢⱥᩥࡘ࠸ ࡚ࡣࠊᅛ᭷ྡモࢆ㝖ࡁࠊయࢆᑠᩥᏐࡋࠊᩥᏐ࡛ጞ ࡲࡿ༢ㄒࡀᅛ᭷ྡモࡋุ࡚᩿ࡉࢀࡿࡇࢆ㜵ࡄࡇ ࡋࡓࠋ ࡑࡢᐇ⾜ࡋ࡚ࠊࢱࢺࣝ⾜ࡀඛ㢌ᩥᏐࡢሙྜ ࠊᑠᩥᏐࡋࡓሙྜࡢTreeTaggerࡢᐇ⾜⤖ᯝࡢ㐪࠸ࢆ ᅗ♧ࡍࠋᅗࡢᕥഃࡣࠊ༢ㄒࡢඛ㢌ࡀᩥᏐࡢሙྜ ࡢ ධ ຊ ࢹ ࣮ ࢱ ࠊ TreeTagger ࡢ ゎ ᯒ ⤖ ᯝ ࡛ ࠶ ࡿ ࠋ “Moments”ࡸ“Systems”ࡢရモࡣࠊᅛ᭷ྡモࢆ♧ࡍNPࡸ ⾲ ༢ㄒศᯒ⾲㑅ู⤖ᯝࡢ୍㒊 ۔㸸㔜せ༢ㄒࠊۻ㸸୰Ꮫ᪤⩦ㄒࠊڹ㸸ࡑࡢ ৭શ ટ >@⋇8VLQJOHWWHUVDQGFKDUDFWHUVLVLPSRUWDQWIRUFRPPXQLFDWLRQ 8VLQJ XVH ۻ OHWWHUV OHWWHU س س ۔ۻ FKDUDFWHUV FKDUDFWHU س ٺ ٺ س س س LPSRUWDQW LPSRUWDQW FRPPXQLFDWLRQFRPPXQLFDWLRQ ٺ ٺ ⋈0DQ\RIWKHPZHUHFUHDWHGWKURXJKXQLTXHSURFHVVHVWKURXJKRXWKLVWRU\ 0DQ\ PDQ\ ۔ FUHDWHG FUHDWH ٺ ٺ س س س ۻ WKURXJK WKURXJK ۔ XQLTXH XQLTXH س س س ۔ SURFHVVHV SURFHVV ٺ ٺ ٺ س ٺ س ۔ WKURXJKRXW WKURXJKRXW س س KLVWRU\ KLVWRU\ س ⋉)RUH[DPSOHn$|LQWKH(QJOLVKDOSKDEHWFRPHVIURPWKHVKDSHRIDQR[KHDG H[DPSOH H[DPSOH (QJOLVK (QJOLVK ڹ DOSKDEHW DOSKDEHW ڹ FRPHVIURPf FRPHV FRPH IURP IURP ۔ۻ VKDSH VKDSH س س س س س ڹ R[ R[ س KHDG KHDG س س ⋊,Q-DSDQHVHKLUDJDQDZDVVLPSOLILHGIURP&KLQHVHFKDUDFWHUV -DSDQHVH -DSDQHVH KLUDJDQD KLUDJDQD VLPSOLILHG VLPSOLILHG ڹ VLPSOLI\ VLPSOLI\ ٺ ٺ ڹ &KLQHVH &KLQHVH ڹ &KLQHVHFKDUDFWHUV &KLQHVHFKDUDFWHUV >@⋇5RRWVRIPDQ\OHWWHUVDQGFKDUDFWHUVHYHQGDWHEDFNWRDQFLHQWWLPHV ۔ URRWV URRW ٺ س س ۔ HYHQ HYHQ س س ٺ س 7UHH7DJJHU 岜ஶୁীෲਛ崟崡崮嵈 岜س؟ৄল峁ୁ岝ٺ؟ୣে嵣ঢ়৴ୁ ஶધীસ崿嵕崘嵑嵈 ઞ৷ୁ ੦ম 1R ୁম ৄলୁ ൕൗਯ ⾲ ༢ᩘᙧ」ᩘᙧࡢぢฟࡋㄒᥖ㍕ᩘ س؟ৄল峁ୁ ৄলୁ ൕൗਯ V\VWHPV PRPHQWV V\VWHP س س PRPHQW س س س ਯ 1R ୁম ୁ ളਯ ரథ ሗ⛉Ꮫⓗࣉ࣮ࣟࢳࡼࡿⱥㄒ᳨ᐃᩍ⛉᭩ࡢㄒᙡศᯒࠉࠉࡑࡢ」ᩘᙧࡢNPSࡀ⾲♧ࡉࢀ࡚࠸ࡿࡓࡵࠊㄒᑿࡢ“s”ࡀ ๐㝖ࡉࢀ࡚࠸࡞࠸ࠋᅗࡢྑഃࡣࠊ༢ㄒࡢඛ㢌ࢆᑠᩥᏐ ኚ ࡋ ࡓ ධ ຊ ࢹ ࣮ ࢱ ࠊ ࡑ ࡢ ゎ ᯒ ⤖ ᯝ ࡛ ࠶ ࡿ ࠋ “moments”ࡸ“systems”ࡢရモࡣྡモࡢ」ᩘᙧ㸦NNS㸧 ゎᯒࡉࢀࠊ༢ㄒࡢㄒᑿࡢ“s”ࡀ๐㝖ࡉࢀ࡚࠸ࡿࠋ ࣆࣜ࢜ࢻࢆ」ᩘ⏝ࡍࡿ┬␎ᙧ༢ㄒࡢグྕ⨨ ⱥ༢ㄒࡢ୰ࡣࠊ“A.M.”ࡸ“P.M.”ࠊ“B.C.”ࡸ“A.D.”ࡢ ࡼ࠺ࠊࣆࣜ࢜ࢻࢆ」ᩘ⏝ࡍࡿ┬␎ᙧ⾲⌧ࡀᏑᅾࡍࡿࠋ TreeTagger࡛ࡣࠊࡇࢀࡽࡢ⾲⌧ࡣࣆࣜ࢜ࢻ࡛ศࡅࡽࢀࠊ 」ᩘࡢ༢ㄒ࡛࠶ࡿゎᯒࡉࢀࡿࠋ࠼ࡤࠊ“A.M.”ࡣ“A.” “M.”ࡢࡘศࡉࢀ࡚ࡋࡲ࠺ࠋࡇࢀࢆ㜵ࡄࡓࡵࠊ ๓ࣆࣜ࢜ࢻࢆูࡢグྕ㸦“A_M_”㸧⨨ࡁ࠼ࡿ సᴗࡀᚲせ࡛࠶ࡿࠋ ✰ᇙࡵၥ㢟ࡢゎ⟅ࡢᇙࡵ㎸ࡳ ᅇࡢᐇ㦂ࢹ࣮ࢱࡢ⠊ᅖእ࡛ࡣ࠶ࡿࡀࠊTreeTaggerࡢ ゎᯒࡣᩥ⬦ࢆ⏝ࡍࡿࡇࡽࠊᩍ⛉᭩ࡢ₇⩦ၥ㢟ࡸ TOEIC࡞ぢࡽࢀࡿ✰ᇙࡵၥ㢟ࢆྵࡴⱥᩥࡣࠊ㑅ᢥ ⫥ࡸゎ⟅ࢆᇙࡵ㎸ࡴ࡞ࡢฎ⌮ࡀᚲせ࡞ࡿࠋ
㸳㸬ࡲࡵ
ᮏ◊✲ࡢ┠ⓗࡣࠊⱥᩥࢆᵓᡂࡍࡿ༢ㄒࢆࠊሗฎ⌮ ᢏ⾡ࢆ⏝࠸࡚ຠ⋡ⓗศᯒࡍࡿ᪉ἲࢆぢฟࡍࡇ࡛࠶ࡗ ࡓࠋࡑࡇ࡛ࠊExcelᶆ‽ഛࡉࢀ࡚࠸ࡿVBAࢆά⏝ࡋࠊ ⱥᩥࢆ༢ㄒศࡍࡿࣉࣟࢢ࣒ࣛࢆ㛤Ⓨࡋࡓࠋศࡉࢀ ࡓ༢ㄒࢆᇶᮏᙧᡠࡍฎ⌮ࢆ⮬ືࡍࡿ㝿ࡣࠊᙧែ⣲ ゎᯒࢶ࣮ࣝࡢWindows∧TreeTaggerࢆ⏝ࡋࡓࠋࡑࡢୖ ࡛ࠊࡇࡢࣉࣟࢢ࣒ࣛࢆࠊඛ⾜◊✲࡛ᵓ⠏ࡋࡓㄒᙡࢹ࣮ࢱ ࣮࣋ࢫ㐃ືࡉࡏࡿࡇࢆヨࡳࡓࠋ ᳨ᐃᩍ⛉᭩ࢆ⏝࠸࡚⾜ࡗࡓᐇ㦂ࡢ⤖ᯝࠊ᪂ࡓ㛤Ⓨ ࡋࡓⱥᩥศࣉࣟࢢ࣒ࣛࢆ⏝࠸ࡿࡇ࡛ࠊᙧែ⣲ゎᯒ ࢶ࣮ࣝࡢWindows∧TreeTaggerࢆࡗ࡚࣐ࣞࡋࡓ༢ㄒ ࢆࠊⱥ༢ㄒศᯒ⾲సᡂࢩࢫࢸ࣒㐺⏝ࡍࡿࡇࡀ࡛ࡁࡿ ࡇࡀ☜ㄆ࡛ࡁࡓࠋࡇࢀࡼࡾࠊⱥᩥฟ⌧ࡍࡿ༢ㄒࢆ ሗฎ⌮ᢏ⾡ࢆ⏝࠸࡚▷㛫࡛ຠ⋡ⓗศᯒࡍࡿ᪉ἲࢆ ☜❧ࡍࡿぢ㏻ࡋࢆᚓࡿࡇࡀ࡛ࡁࡓࠋ ᅇࡢ◊✲ᡂᯝࡢ୍ࡋ࡚ࠊᐇ㦂⤖ᯝࢆࡶస ᡂࡋࡓᤵᴗ⏝ࡢࣉࣜࣥࢺᩍᮦࢆᅗ♧ࡍࠋ㸴㸬ᚋࡢㄢ㢟
ᚋࡢㄢ㢟ࡋ࡚ࡣࠊᅇࡢᐇ㦂᪉ἲࡢ⮬ືࢆ᳨ ウࡋ࡚࠸ࡿࠋ࠼ࡤࠊ⌧ᅾࡣࢥ࣐ࣥࢻࣉࣟࣥࣉࢺࡽᡭ ධຊ࡛ᐇ⾜ࡋ࡚࠸ࡿTreeTaggerࡣࠊVBAࣉࣟࢢ࣒ࣛ⤌ ࡳ㎸ࡴࡇࡀ࡛ࡁࡿࠋᮏ◊✲࡛㛤Ⓨࡋࡓⱥᩥศࣉࣟࢢ ࣒ࣛࡢࣝࢦࣜࢬ࣒ࢆࡶࠊExcelୖࡽධຊࢹ࣮ࢱ ࡢᩚᙧࢆ⾜࠸ࠊTreeTaggerࢆᐇ⾜ࡋࠊࡑࡢ⤖ᯝࢆ⮬ື㞟 ィࡍࡿࡲ࡛ࡢ୍㐃ࡢసᴗࢆࠊ⮬ືࡍࡿࡇࢆ㐍ࡵ࡚࠸ ࡁࡓ࠸ࠋ ࡲࡓࠊᮏ◊✲࡛ᵓ⠏ࡋࡓㄒᙡศᯒࡢᡭἲࡣࠊࡉࡲࡊ ࡲ࡞ⱥᩥ㐺⏝ྍ⬟࡛࠶ࡿࠋᙜ㠃ࡢㄢ㢟ࡋ࡚ࡣࠊ ᖺ ᭶ ࡽ ᪂ ᙧ ᘧ ኚ ᭦ ࡉ ࢀ ࡓ TOEIC Reading & Listening Testࡢࠗබᘧၥ㢟㞟࠘㸦Educational㸧ࡢ༢ㄒࢆ ศᯒࡍࡿணᐃ࡛࠶ࡿࠋ/HVVRQ3LFWXUHVRI)XQQ\0RPHQWV OHVVRQSLFWXUHVRIIXQQ\PRPHQWV /HVVRQ:ULWLQJ6\VWHPVLQWKH:RUOG OHVVRQZULWLQJV\VWHPVLQWKHZRUOG
੪峘ୁ ષဨ লৡୁ ੪峘ୁ ષဨ লৡୁ
/HVVRQ 11 OHVVRQ OHVVRQ 11 OHVVRQ
&' &'
3LFWXUHV 116 SLFWXUH SLFWXUHV 116 SLFWXUH
RI ,1 RI RI ,1 RI
)XQQ\ 13 )XQQ\ IXQQ\ -- IXQQ\ 0RPHQWV 13 0RPHQWV PRPHQWV 116 PRPHQW /HVVRQ 13 /HVVRQ OHVVRQ 11 OHVVRQ
&' &'
:ULWLQJ 13 :ULWLQJ ZULWLQJ 99* ZULWH 6\VWHPV 136 6\VWHPV V\VWHPV 116 V\VWHP
LQ ,1 LQ LQ ,1 LQ
WKH '7 WKH WKH '7 WKH
:RUOG 13 :RUOG ZRUOG 11 ZRUOG
ୁ峘୍পધஊ ୁ峘୍৵ધஊ োৡஶધ 7UHH7DJJHU峘ৰষટ ᅗ ༢ㄒࡢඛ㢌ᩥᏐࡢ㐪࠸ࡼࡿゎᯒ⤖ᯝࡢ┦㐪 ᅗ ᤵᴗࣉࣜࣥࢺᩍᮦ 1Rقك 崗嵑崡 ତ৶ಀ ৾ආಀ /HVVRQ :ULWLQJ6\VWHPVLQWKH:RUOG 6HFWLRQ 'R\RXUHDGRUZULWHVRPHWKLQJHYHU\GD\"*<RXSUREDEO\HQMR\ UHDGLQJERRNV<RXPD\VRPHWLPHVZULWHHPDLOVWR\RXUIULHQGV
*8VLQJ OHWWHUV DQG FKDUDFWHUV LV LPSRUWDQW IRU FRPPXQLFDWLRQ 0DQ\ RI WKHP ZHUH FUHDWHG WKURXJK XQLTXH SURFHVVHV WKURXJKRXW KLVWRU\ )RUH[DPSOH“$”LQWKH(QJOLVKDOSKDEHWFRPHVIURPWKH VKDSH RI DQ R[ KHDG ,Q -DSDQHVHKLUDJDQD ZDV VLPSOLILHG IURP &KLQHVHFKDUDFWHUV
5RRWVRI PDQ\OHWWHUVDQGFKDUDFWHUV HYHQGDWH EDFNWR DQFLHQW WLPHV7KHLUVKDSHVKDYHFKDQJHGRYHUWLPH ୁ嵣࿃ୁ ஶધর峑峘ਔ峼ോછ峒ઇఐછ峼৹峣峐છ岷峔岿岮岞 ٵ峙ਏୁ ٲ峙র৾ใಆୁ V\VWHPV ٵSUREDEO\ . ٲHQMR\ ٲOHWWHU 峕ఠધஊ ٵٲFKDUDFWHU 峕ਔધஊ ٵFUHDWHG . ٲWKURXJK ٵXQLTXHЍ/HVVRQ . ٵSURFHVVHV ٵWKURXJKRXW . DOSKDEHW FRPHVIURP㹼 . ٵٲVKDSH R[ . VLPSOLI\ &KLQHVH . &KLQHVHFKDUDFWHUခஊ ٵURRWV . ٵHYHQ GDWHEDFNWRaع峕岿岵峘峦峵 ٵDQFLHQWЍ/HVVRQ RYHUWLPH ৎ岶峉峎峕峎島 ધ 峘ஶધ峕 6岝9岝&岝2 峘੶ಀ峼੶ো峁峔岿岮岞 ࠉࠉ7+(*810$̻.2+6(15(9,(:1R
ᮏ◊✲࡛㛤Ⓨࡋࡓⱥᩥศࣉࣟࢢ࣒ࣛࢆ࠺ࡇ࡛ࠊ ධຊࢹ࣮ࢱࡢᩚᙧࢆࡋࡓࢡࣞࣥࢪࣥࢢసᴗࡸࠊฟ⌧ ࡍࡿ༢ㄒᩘࡢ㞟ィసᴗࢆ⮬ື࡛⾜࠺ࡇࡀ࡛ࡁࡿࠋࡲࡓࠊ TreeTagger࡛࣐ࣞࡋࡓ༢ㄒࢆࠊⱥ༢ㄒศᯒ⾲సᡂࢩࢫ ࢸ࣒㐺⏝ࡍࡿࡇ࡛ࠊ༢ㄒࢆฟ⌧㢖ᗘࡢ್ࡽᶵᲔⓗ 㑅ูࡋࠊ㔜せ༢ㄒࣜࢫࢺࢆసᡂࡍࡿࡇࡶྍ⬟࡛࠶ࡿࠋ ࡇ࠺ࡋࡓᇶ♏ⓗ◊✲ࢆ⥅⥆ࡋ࡞ࡀࡽࠊᏛ⩦⪅ࡀࡼࡾ ຠᯝⓗㄒᙡࢆ⩦ᚓ࡛ࡁࡿ᪉ἲࢆ᥈ồࡋࠊࡑࡢࡓࡵࡢᏛ ⩦⎔ቃࡢᩚഛࢆ⾜ࡗ࡚࠸ࡁࡓ࠸ࠋ ཧ⪃ᩥ⊩
Educational Testing Service (2016).ࠗTOEICࢸࢫࢺබᘧၥ 㢟㞟㸸᪂ᙧᘧၥ㢟ᑐᛂ⦅࠘ᮾி㸸ᅜ㝿ࣅࢪࢿࢫࢥ ࣑ࣗࢽࢣ࣮ࢩࣙࣥ༠㸬
Schmid, Helmut. (1994). “Probabilistic Part-of-Speech Tagging Using Decision Trees.” Proceedings of
International Conference on New Methods in Language Processing, 44-49.
㸫㸫㸫“TreeTagger.” Retrieved August 8, 2016, from: http: //www.cis.uni-muenchen.de/~schmid/tools/TreeTagger UNIVERSITY of WASHINGTON (UW) Courses Web
Server. “Tree Tagger Tags.” Retrieved August 8, 2016,
from: https://courses.washington.edu/hypertxt/csar-v02/ penntable.html ᒾᓮὒ୍ࠕࢩ࢙ࣝࢫࢡࣜࣉࢺࢆά⏝ࡋࡓⱥㄒㄒᙡศᯒ ࣉࣟࢢ࣒ࣛࡢᣑᙇࠖࠗᮌ᭦ὠᕤᴗ㧗➼ᑓ㛛Ꮫᰯ⣖ せ࠘㸸 ⓑᜤᘯࠗእᅜㄒᏛ⩦ࡢ⛉Ꮫ̿̿➨ゝㄒ⩦ᚓㄽࡣ ఱ࠘ᒾἼ᪂᭩ᮾி㸸ᒾἼ᭩ᗑ ⏣୰┬స㸬ࠕᙧែ⣲ゎᯒࢶ࣮ࣝ̿ⱥㄒ TreeTagger ࢆ ୰ᚰ̿ࠖࠗᕞᏛሗᇶ┙ࢭࣥࢱ࣮ᗈሗ࠘㸬 9RO1R㸸㸬 ⏣୰┬సࠊᾆὒ୍ࠊᚨぢ㐨ኵ㸬ࠕᶵ㛵࣏ࣜࢪࢺࣜࡽ ᚓࡽࢀࡿⴭ⪅ࡢㄒᙡศᕸᇶ࡙࠸ࡓ㒊ᒁู㔜せㄒᙡ ࡢ㑅ᐃࠖࠗேᩥ⛉Ꮫࢥࣥࣆ࣮ࣗࢱࢩ࣏ࣥࢪ࣒࢘࠘㸬 ᖺ1R㸸 ඵ㫽ྜྷ᫂ࠊ㔝ె௦Ꮚࠕⱥ༢ㄒᏛ⩦⏝ Web ᩍᮦ㛤Ⓨ ̿̿ຠᯝⓗ࡞ⱥㄒᏛ⩦ἲࢆồࡵ࡚ࠖࠗ⩌㤿㧗ᑓࣞ ࣅ࣮ࣗ࠘㸸 㸫㸫㸫ࠕⱥ༢ㄒ㑅ᢥࢩࢫࢸ࣒ࡢ㛤Ⓨᛂ⏝ࠖࠗ⩌㤿㧗 ᑓࣞࣅ࣮ࣗ࠘㸸 㸫㸫㸫ࠕ㧗ᑓ⏕ᚲせ࡞ㄒᙡࢆồࡵ࡚̿̿Ꮫ⩦⏝ㄒᙡ ࣜࢫࢺࢆ⏝࠸ࡓㄒᙡࢹ࣮ࢱ࣮࣋ࢫࡢᵓ⠏ࡑࡢ᳨ド ̿̿ࠖࠗ&2&(7◊✲ㄽ㞟࠘㸸 ᳃ఫ⾨ࠊ ྡࠗMY WAY English Communication I࠘
ᮾி㸸୕┬ᇽ
Analyzing the Vocabulary
of a Certified English Textbook:
An Information-Technology Approach
Yoshiaki HACHITORI and Kayoko OHNO
This study examines a method we constructed to analyze words used to form sentences in the English language by applying information-processing technology. In order to develop effective methods to increase students’ vocabulary, we have been conducting research on introducing information-processing technology in English language education. In this study, we developed a program that deconstructs sentences in English into words using the Visual Basic for Applications (VBA) of Excel as the programming language. TreeTagger, a morphological analysis tool, was used to render words in their basic form. Linking this program to our vocabulary database, which was developed in a separate study, enabled us to choose the important words based on frequency analysis. To evaluate the efficiency of this method, an experiment was conducted using a certified English textbook. The results of the experiment show that this method is effective in analyzing different parts of speech and frequency of word usage in sentences, and the analysis takes a short amount of time.