ύЎཥᆪϐᜢᖄຒᙚ
ᖰ මర
ЎϯεᏢၗᆅ܌ୋ௲! ЎϯεᏢၗᆅ܌ࣴزғ
cshwang@faculty.pccu.edu.tw
! 92707157@scenet.pccu.edu.tw
ᄔा
Ӣࣁᆛሞᆛၡ٬ҔຫٰຫεϯǴаϷࣽמ מೌޑǴၗૻሀຫٰຫܰǴၗૻໆΨВ ቚуǴӧεໆޑၗૻύǴ٬Ҕޣຫٰຫ֚ᜤפډԖ ҔޑၗૻǶ ҞεҽၗૻᔠޑБԄǴࢂၸӄЎཛྷ ൨Ǵཛྷ൨р಄ӝ٬Ҕޣ܌ᒡΕᜢᗖຒޑၗǴ၌ ޑೲࡋߚதᄌǴԶЪ٬Ҕޣӵ݀ᒡΕϼቶݱ܈ᒱ ᇤޑᜢᗖຒǴᗋࢂᒪᅅӭၗૻǴࣁΑ෧Ͽ೭ ኬޑୢᚒǴ߾Ѹሡவຒᆶຒϐ໔ޑᜢᖄ܄ЋǴԾ ϯࡌᄬᜢᖄຒڂǶҁࣴز२ӃճҔᜪઓᆛၡᄽ ᆉݤǴԾᘏڗрϪӝЎཀޑᜢᗖຒǴӆаຒᓎϸ ᙯᓎޑख़ीᆉϦԄǴीᆉрᜢᗖຒϐ໔ޑᜢᖄ ख़Ǵࡌᄬޔௗ܈໔ௗᜢᖄޑຒڂǴ٠Ъගٮ٬Ҕ ޣᘤំ܈ᔠޑୖԵǶ ҁࣴزᒿᐒᒧڗᖄӝཥᆪᆛޑ 500 ጇཥᆪЎ ҹǴჴᡍ่݀ᡉҢѳ֡ጇཥᆪԖ 3 ঁࢂ಄ӝΓπ ۓကᜢᗖຒǴځᎩᗨฅό಄ӝΓπۓကǴՠऊϤԋ ё٬Ҕޣ܌ௗڙǶჹጇЎകǴךॺᘏڗ 20 ঁᜢᗖຒࣁж߄ຒ༼Ǵٰीᆉځᜢᖄ܄Ƕ ᜢᗖຒǺᜢᗖຒǵᜪઓᆛၡǵॹሀኳࠠǵ໔ௗ ᜢᖄຒǶ1.
ᆣፕ
җܭᆛሞᆛၡޑጲࠁวǴаϷၗૻࣽמВ ǴԖຫٰຫӭޑၗૻӧᆛၡሀǴჹ٬Ҕޣ ٰᇥǴाڗளၗૻࢂߚதܰϷБߡޑǴՠࢂӵՖ ӧεໆޑၗૻύǴᘏڗԖҔޑၗૻࢂঁ࣬ख़ा ޑፐᚒǶ ҞӚεᆛઠගٮޑཛྷ൨БԄǴεҽࢂගٮ ᜢᗖຒՉཛྷ൨Ǵٯӵ٬Ҕޣགྷפکًؓౢޑ࣬ ᜢၗૻǴᒡΕᜢᗖຒȸًؓȹǴջрЎകύ ϣ֖ȸًؓȹᜢᗖຒޑЎകǴӵ݀٬ҔޣԖᑫ፪ځ ύጇЎകǴޔௗᗺᒧϷ᎙᠐၀ጇЎകǴӵ݀၀ ጇЎക٠όࢂ٬ҔޣགྷाޑǴ߾Ѹख़ཥӆ၌ځ дᜢᗖຒޑЎകǶᗨฅ೭ঁБݤςှ،٬Ҕޣό ሡाጇጇЎകᘤំǴջёפрሡाޑၗૻǴՠࢂ ӵ݀٬ҔޣᒡΕϼቶݱ܈ᒱᇤޑᜢᗖຒǴᗋࢂคݤ ᔅշ٬ҔޣǴפډወӧԖҔޑၗૻǶ ጇЎകࢂҗӭຒ༼܌ಔԋޑǴӧ೭٤ຒ༼ ӝύԖ٤ख़ाຒ༼ࢂёаᘏڗрٰǴԋอ Ўᄔा܈ࢂЎകޑЇຒǴ೭٤ख़ाޑຒ༼Ǵ ᆀϐࣁȨᜢᗖຒȩǴΨ൩ࢂӄЎख़ᗺᜢᗖޑຒ༼Ƕ ᜢᗖຒϐ໔ޑᜢᖄ܄ղۓǴࢂаΒຒӅӕр ޑᓎࣁЬǴٯӵᜢᗖຒȸျໜȹکȸRFIDȹӅӕ рӧЎҹύޑᓎࡐଯǴ܌аёղᘐځԖޔௗᜢ ᖄǹԶȸ҉ᙦᎩȹکȸRFIDȹӕਔрޑᓎΨࡐ ଯǴղᘐΨԖޔௗᜢᖄǴ߾ȸျໜȹǵȸRFIDȹϷ ȸ҉ᙦᎩȹ೭ΟঁᜢᗖຒԖᜢᖄǶӢԜȸျໜȹϷ ȸ҉ᙦᎩȹԖ໔ௗᜢᖄǴ٬Ҕޣёၸᜢᗖຒޑό ӕᜢᖄǴפډӕሦୱࠅόӕБӛޑၗૻǶ ՠࢂӧεໆޑЎӷϐύǴΓπᘏڗрᜢᗖຒϷ ᜢᖄຒࢂߚதਔ໔ϷΓΚǴӢԜҁࣴزගрԾ ϯᘏڗрᜢᗖຒǴаϷԾࡌᄬᜢᖄຒ༼Ƕ ٬Ҕޣཛྷ൨ၗૻਔǴନΑගٮЎകύϪӝЎཀޑ ᜢᗖຒǴӕਔΨᙚԖޔௗ܈໔ௗ࣬ᜢޑᜢᖄຒǶ ҁࣴزճҔᜪઓᆛၡޑᄽᆉݤǴࡌҥԾᘏ ڗрᜢᗖຒޑ૽ግኳࠠǴӆаຒᓎϸᙯᓎޑख़ ीᆉϦԄǴीᆉрᜢᗖຒϐ໔ޑᜢᖄख़Ǵڗளޔ ௗϷ໔ௗᜢᖄຒǶ2.
Ў
2.1 ᜢᗖຒᘏڗ ᜢᗖຒࢂЎകԖཀကޑനλಔԋൂՏǴεҽ ޑЎҹԾϯೀǴٯӵԾᄔाǵԾЇϷԾ ϩᜪǴӃᜢᗖຒᘏڗբǴӆՉࡕុ ೀǶёаᇥǴᜢᗖຒᘏڗࢂ܌ԖЎҹԾೀޑ ୷ᘵᆶਡЈמೌǶ ᜢᗖຒᘏڗޑБݤǴёεౣϩࣁीݤǵຒ ݤǵೕ߾ݤ܈೭ΟᅿБݤޑӝٳၮҔǶӧၸѐޑЎ ύǴᜢᗖຒᘏڗޑמೌёаϩԋΟεᜪࠠ [13] [14]Ƕ ಃᅿࣁຒКჹݤǺջճҔςࡌҥޑຒǴ ٰКჹᒡΕЎҹȐ܈ЎѡȑǴᘏڗЎҹύр಄ӝ ຒύޑТᇟǶ ಃΒᅿࣁЎݤওݤǺၸԾฅᇟقೀמೌ ޑЎݤওำԄǴওрЎҹύޑӜຒТᇟǴӆၮ Ҕ٤Бݤᆶྗ߾Ǵၸᘠόӝޑຒ༼Ƕ ಃΟᅿБݤࣁीϩݤǺၸჹЎҹޑϩ ǴಕᑈىޑीୖኧࡕǴӆஒीୖኧ಄ӝߐ ᘖॶޑТᇟᘏڗрٰǶ ځдޑБݤᗋхࡴॊБݤޑᆕӝၮҔǴ܈ၮ ҔόӕޑᄽᆉݤǶٯӵ Krulwich, B. and Burkey, C.[3] ࣁΑЎകԾϩᜪǴճҔᡍݤ߾ᄽᆉݤǴவ
ЎകύᘏڗрᜢᗖຒǴբࣁϩᜪޑቻॶǴฅԶځ ჴᡍ่݀ࠅᘏڗрεໆЪեᆒዴࡋޑᜢᗖຒǶ
Muñoz, A. [4] ගрคᅱ࿎ԄᏢಞБݤٰᘏڗ
Β ঁ ӷ ޑ ᜢ ᗖ ຒ Ǵ ௦ Ҕ Ծ ᔈ Ӆ ਁ ፕ ᆛ ၡ
(Adaptive Resonance TheoryNetwork, ART)Ǵځ่݀
ΨࢂᘏڗрεໆЪեᆒዴࡋޑᜢᗖຒǶSteier, A. M.,
and Belew, R. K. [8]٬Ҕ࣬ϕૻ৲ڄኧ (Mutual
information) ٰीᆉᜢᗖຒቻॶǴՠځБݤѝૈ
ௗڙΒঁӷޑᜢᗖຒǶ
Turney, P.D. [9][10] ගр Genex ࢎᄬǴЬाа
ᒪ୷Ӣᄽᆉݤ (GeneticAlgorithm ,GA)ᘏڗᜢᗖ ຒǴ่݀ѳ֡ጇЎകᘏڗрΒঁᜢᗖຒǶ Witten,
I.H., Paynter, G.W., Frank, E., Gutwin[12] ගрঁ KeaჴբࢎᄬǴ٬ҔنԄ(Bayesian)ᄽᆉݤǴԜᄽᆉ
ݤӧ Turney, P.D.[11]ύჴᡍܴр Kea ک Genex Ԗ εऊ࣬ޑਏǶ 2.2 ຒ༼ख़ीᆉ ӧၗૻᓯӸᆶᔠޑጄᛑԶقǴЇᜏڂࢂ ᒵຒ༼ϐ໔໘ቫ܈ᇟཀޑᜢ߯Ǵࣁ٬Ҕޣᔠၗ ਔǴёၸЇᜏڂᙚ࣬՟ཷۺޑӷ܈ຒǶ ЇᜏڂࢂᒵӕကຒǴᗋԖϸကຒǵቶ ကຒǵကຒǵ࣬ᜢຒǴҔаᘉ܈ᕭλᔠຒ ༼ޑЬᚒጄൎǶຒཀ࣬ᜢޑЇᜏڂǴѸҗΓΚ ᆢៈǴЎകኧໆຫٰຫӭǴ߾ाԖ׳ӭޑΓΚϷ ਔ໔ωૈᆢៈЇᜏڂǶࣁΑૈԖਏϷԾϯࡌ ҥЇᜏڂǴךॺаຒ༼Ӆӕрޑᜢ߯Ǵٰࣁ ຒ༼ϐ໔ޑᜢᖄǶ
Aas, K. and Eikvil, L.[1]рӚᅿόӕຒ༼
ख़ޑीᆉϦԄǴճҔຒ༼ख़ٰຑЎകޑ࣬՟ ࡋǴӵΠॊǺ NࢂࡰЎҹޑᕴኧҞǴM ࢂࡰᘐຒࡕޑຒ༼ᕴ ኧǴniࢂࡰຒ༼ i рޑЎҹኧǶ )2* BooleanǺനᙁൂޑीᆉБԄǴӵ݀၀ຒ༼р ӧ೭ጇЎകǴ߾ख़ॶࣁ 1Ǵϸϐ߾ࣁ 1Ƕ! fikࢂࡰຒ༼ i рӧЎക k ύޑԛኧǶ ⎩ ⎨ ⎧ > = otherwise 0 0 if 1 ik ik f w
(2) word frequency weightingǺຒ༼ i рӧЎക k
ύޑԛኧǶ ik ik f w = (3) TF
×
IDF weightingǺຒᓎϸᙯЎҹᓎǶ ) log( * i ik ik n N f w = (4) tfc-weightingǺԵቾډόӕޑЎകߏࡋǴीᆉ ၀ຒޑຒᓎϸᙯЎҹᓎӧЎകύޑКٯǶ∑
= = M j k j j i ik ik n N f n N f w 1 2 )] log( * [ ) log( * (5) ltc-weightingǺᜪ՟ tfc-weightingǴࣁᗉխଯຒ ᓎޑቹៜǴԶፓຒᓎǶ∑
= + + + = M j j jk i ik ik n N f n N f w 1 2 )] log( ) 0 . 1 [log( ) log( * ) 0 . 1 log( (6) EntropyǺ⪖ख़Ǵࢂঁፄᚇޑख़ीᆉБ ݤǴǶ ⎟⎟⎠ ⎞ ⎜⎜⎝ ⎛ + + = ∑ = ) log( ) ) log( 1 1 ( * ) 0 . 1 log( 1 i ij N j i ij ik ik n f n f N f w ॊ܌ගрޑຒ༼ख़ीᆉϦԄǴЬाࢂࣁ ЎകϩᜪਔǴीᆉЎകޑ࣬՟ࡋǶҁࣴزճҔ೭٤ ຒ༼ख़ϦԄǴٰीᆉຒ༼໔ӧύЎཥᆪЎകύޑ ࣬՟ࡋǶ 2.3 ᜪઓᆛၡ ᜪઓᆛၡࢂᅿኳᔕғނઓسޑೀ سǶғނઓسҗӭઓϡ࣬ϕೱ่ǴԶ ঁઓϡԖᒡрϷᒡΕૻဦکځдઓϡ࣬ೱ Ϸሀ৲ǶҞᜪઓᆛၡёϩࣁΠᜪΟᅿǺ ಃᅿᅱ࿎ԄᏢಞᆛၡǴவୢᚒሦୱύගٮ૽ ግጄٯǴх֖ᒡΕၗϷᒡрၗǶ٠Ъவᆛၡύ ᏢಞᒡΕၗᆶᒡрၗޑϣӧჹࢀೕ߾ǶதᔈҔ ܭႣෳ܈ϩᜪǶٯӵچϩભ[2]کઇౢႣෳ[6]Ƕ ҁࣴز܌٬Ҕޑॹሀઓᆛၡ(Back -Propagation Network) ջࢂឦܭԜᜪࠠǶ ಃΒᅿคᅱ࿎ԄᏢಞᆛၡǴவୢᚒሦୱύڗள ѝԖᒡΕၗޑ૽ግጄٯǴ٠வᆛၡύᏢಞᒡΕၗ ޑϣӧᆫᜪೕ߾ǴаᔈҔܭཥޑਢٯǶٯӵԾಔ ᙃࢀკᆛၡ(Self-Organizing MapǴSOM)ǵԾᔈ Ӆ ਁ ፕ ᆛ ၡ (Adaptive Resonance TheoryNetworkǴART) Ƕ
ಃΟᅿᖄགྷԄᏢಞᆛၡǴаރᄊᡂኧॶࣁ૽ግ ጄٯǴᏢಞጄٯύޑᏫೕ߾ǴฅࡕᔈҔܭѝԖό ֹރᄊॶǴԶሡፕֹރᄊޑཥਢٯǴ೭ᅿᆛ ၡёаᔈҔܭᘏڗᔈҔᆶᚇૻၸᘠǶٯӵᓅදߚᅟ ᆛၡ(Hopfield Neural Network)аϷᚈӛᏫᆛၡ
(Bi-directional Associative Memory)ឦϐǶ
3.
سࢎᄬ
ҁࣴزسࢎᄬϩԋΒঁኳಔǴಃኳಔࢂԾ ϯᘏڗрᜢᗖຒ༼ǴಃΒኳಔࢂճҔঁኳಔ ޑᜢᗖຒ༼Ǵीᆉຒ༼໔ޑ࣬՟ࡋǴࡌҥᜢᖄຒڂǶ
3.1 Ծϯᜢᗖຒᘏڗ ᜢᗖຒᘏڗᡯඔॊӵΠǺ (1) ࡌҥ૽ግኳࠠǺӃҗΓπБԄٰۓက૽ ግЎҹޑᜢᗖຒǴӆճҔۓကрޑᜢᗖ ຒٰࡌҥঁ૽ግኳࠠǶ࣬ᜢࢬำፎـ კǶ (2) ᘏڗᜢᗖຒǺճҔᡯޑኳࠠǴᘏ ڗрෳ၂ЎҹύޑᜢᗖຒǶ࣬ᜢࢬำፎ ـკΒǶ ߄ 1 ຒ܄ೕ߾ޑၸᘠ 1. Ӝຒ 2. ຒɠӜຒ 3. ӜຒɠӜຒ 4. Ӝຒɠຒ 5. ຒɠӜຒ ύЎᘐຒࢂճҔύࣴଣޑύЎᘐຒس[5]ٰ ᘐຒǴӆஒᘐрޑຒ༼ၸຒ܄ೕ߾ၸᘠǴᘏڗ рংᒧຒǴӵ߄ 1Ƕ ॊբᘏڗрٰޑংᒧຒǴѸӆीᆉΟ ቻॶǴࣁ૽ግၗޑቻǴӵ߄ 2Ƕ೭Ο૽ ግቻॶޑඔॊӵΠǺ (1) ຒ༼рޑख़ǺҁӢનԵቾຒ༼ӧጇЎ കрޑՏόӕǴԶԖόӕޑख़ा܄Ǵ܌ аຒ༼рόӕޑՏǴ߾Ԗόӕޑख़Ƕ ٯӵрӧܩᓐǴ߾ख़ࣁ w1Ǵрӧಃ ࢤǴ߾ख़ࣁ w2ǴځдӦБޑख़ࣁ w3Ǵी ᆉБݤӵϦԄ(1)܌ҢǶځύ 1 ik f ࢂຒ༼ i ӧЎ ҹ k ύǴрӧܩᓐޑຒᓎǴ 2 ik f ࢂຒ༼ i ӧ Ўҹ k ύǴрӧಃࢤޑຒᓎǴ 3 ik f ࢂຒ༼ iӧЎҹ k ύǴрӧځдՏޑຒᓎǶ 3 3 2 2 1 1 w f w f w f PWik = ik× + ik× + ik× (1) (2) ࣬ჹຒߏǺ߄Ңຒ༼ߏࡋନаЎകύ܌Ԗຒ༼ ޑѳ֡ߏࡋǶ (3) TFØIDFǺຒᓎϸᙯЎҹᓎǴԖٿঁ୷ҁଷ ǺঁຒрӧҽЎҹύԛኧຫӭ߾ຫख़ ाǹऩӧ܌ԖᇆЎҹύрԛኧຫӭ߾ຫό ख़ाǴӢࣁ߄Ң೭ຒคݤж߄೭ҽЎҹޑ ܄ǴځीᆉБݤӵϦԄ(2)܌ҢǴځύ fikࢂຒ ༼ i ӧЎҹ k ޑຒᓎǴN ࣁᕴЎҹኧǴniࣁԿ Ͽрԛຒ༼ i ޑЎҹኧǶ ) log( * i ik ik n N f TFIDF = (2) კ 1 ࡌҥ૽ግኳࠠ კ 2 ᘏڗᜢᗖຒ
߄ 2 ૽ግၗޑቻ ቻӜᆀ ඔॊ ຒᓎ ຒ༼рӧܩᓐǵ२ࢤϷځдՏޑ ຒᓎуᕴکǶ ࣬ჹຒߏ ຒ༼ޑߏࡋନаЎകύޑ܌Ԗຒ༼ ѳ֡ߏࡋǶ TFØIDF ຒ༼ޑ TFØIDF ॶǶ ӕਔҗৎΓπۓကጇཥᆪޑᜢᗖຒǴӆஒ ϐύЎᘐຒسᘏڗрٰޑংᒧຒ༼Ǵၸᘠߏ ࡋλܭ 2 ޑຒ༼Ǵीᆉॊ૽ግቻǴуࢂցࣁ ᜢᗖຒǴԋ૽ግၗǴၸᜪઓᆛၡޑॹ ሀБݤ૽ግၗǴࡌҥрᜢᗖຒޑ૽ግኳࠠǶ ಃΒᡯࢂаځдཥᆪЎҹբෳ၂ǴӃճҔύ ЎᘐຒسᘏڗрংᒧຒǴीᆉঁຒ༼ޑॊΟ ቻǴճҔಃᡯޑᜢᗖຒ૽ግኳࠠǴٰᘏڗ рᜢᗖຒǶ 3.2 ᜢᖄຒᙚ ᜢᖄຒᙚᡯඔॊӵΠǺ (1) ीᆉຒ༼ख़ǺஒᘏڗрޑᜢᗖຒǴी ᆉځӧጇЎകޑख़Ǵࡌҥຒ༼ख़ ᔞǶ (2) ीᆉຒ༼࣬՟ࡋǺஒॊᡯޑຒ༼ ख़ᔞǴीᆉٿٿຒ༼ӧෳ၂Ўകύޑ࣬ ՟ࡋǴᘏڗрԖᜢᖄޑຒ༼Ƕ ॊ࣬ᜢࢬำፎـკ 3Ƕ კ 3 ᜢᗖຒᙚ ཥᆪޑϣ٠όߏǴаϷຒᓎΨόଯǴ܌аҁ ࣴ ز ௦ Ҕ Aas,K. and Eikvil, L.[1] ܌ ග р ޑ
tfc-weighting БݤǴीᆉӚຒ༼ӧෳ၂ཥᆪЎകύ ޑख़ǴीᆉϦԄӵ(3)ǴN ࢂࡰЎҹޑᕴኧҞǴM ࢂࡰᘐຒࡕޑຒ༼ᕴኧǴniࢂࡰຒ༼ i рޑЎҹ ኧǶ
∑
= = M j k j j i ik ik n N f n N f w 1 2 )] log( * [ ) log( * (3) ीᆉֹຒ༼ख़ࡕǴҁࣴزаӛໆޜ໔ٰ߄Ң ЎകϷຒ༼܌ᄬԋޑΒᆢࡋޜ໔ǴٯӵԖ n ጇཥᆪ ЎകǴϷ܌ԖЎകύᕴӅрԖ m ঁຒ༼Ǵ߾ࡌҥ n*mޑΒᆢࡋޜ໔Ǵӵკ 4Ƕ аΒᆢࡋޜ໔ٰीᆉຒຒ༼໔ޑᜢᖄ܄Ǵᜢᖄ ܄ෳໆБԄа Salton,Gerard [7]ගрޑӛໆᜢᖄ܄ ीᆉϦԄǴӵ(4) Ǵwikࢂࡰຒ༼ i ӧЎക k ύޑ ख़Ƕ∑
= × = n k jk ik j i T w w T sim 1 ) , ((4) კ 4 ӛໆޜ໔
4.
ჴբ่݀
ҁࣴزаᖄӝཥᆪᆛ 11~12 ДޑཥᆪǴᒿ ᐒᒧڗ 500 ጇࣁࣴزჹຝǴϩԋ 400 ጇࣁ૽ግЎ ҹϷ 100 ጇࣁෳ၂ЎҹǶ ҁࣴزύޑಃ૽ግቻǴຒ༼рޑ ख़Ǵࣁуख़рӧܩᓐ܈२ࢤޑᜢᗖຒޑख़Ǵ ۓрӧܩᓐǴ߾ख़ࣁ 2ǴрӧಃࢤǴ߾ ख़ࣁ 1.5ǴځдӦБޑख़ࣁ 1Ƕ ԖᜢԾᘏڗᜢᗖຒǴҁࣴز௦ڗΒᅿБԄٰ ຑኳࠠԋਏǴಃᅿࢂ Ian H. Witten Γӧ Kea ኳࠠύ܌ගрޑБݤǴவЎകύᘏڗрᜢᗖຒǴी ᆉԖӭϿࢂ಄ӝΓπۓကޑᜢᗖຒǴځЬाচӢӵ ΠǺ (1) ԜБݤКҔᆒዴϷєӣ׳ܰ٬ ҔޣှǶ (2) ᆒዴϷєӣёૈᇤᏤ٬ҔޣǴࣁ ΑଓଯᆒዴԶ឴࣊ΑєӣǴ܈ଓ ଯєӣԶ឴࣊ΑᆒዴǶ (3) ҁБݤ಄ӝ٬ҔޣதаЎക܌ᘏڗрᜢ ᗖຒኧໆٰᑽໆǶ ߄ 3 ൩ࢂаጇෳ၂Ўകڗр 5ǵ10ǵ15ǵ 20ঁᜢᗖຒǴीԖӭϿঁࢂ಄ӝΓπۓကޑᜢᗖ ຒǴ٠ीᆉጇჴሞ಄ӝޑᜢᗖຒኧໆǶKea ኳࠠ аमЎයтࣁࣴزჹຝǴᘏڗрጇޑ 5ǵ10ǵ15ǵ20 ঁຒ༼ύѳ֡Ԗ 0.93ǵ1.39ǵ1.68ǵ1.88 ঁ ಄ӝΓπۓကޑᜢᗖຒǴᗨฅکҁࣴزޑෳ၂Ўҹ ޑᇟقϷϣόӕǴคݤ࠼ᢀޑКၨǴՠࢂҁࣴز ኳࠠޑჴᡍ่݀ᡉҢК Kea ޑၨ٫Ƕ ߄ 3 ᜢᗖຒኧໆ ᘏڗᜢᗖຒኧໆ ಄ ӝ Γ πۓက 5 10 15 20 ѳ֡ 1.98 2.7 2.99 3.1 ಃΒᅿ߾ࢂаനதـޑᆒዴ(Precision)Ϸє ӣ(Recall)ٰຑǴ่݀ፎـკ 5Ǵी૽ግрჴ ሞ಄ӝᜢᗖຒޑᆒዴϷєӣǴ่݀߄Ңрӧᘏ ڗຒ༼ኧໆຫӭǴ߾єӣຫଯǴᆒዴຫեǶ კ 5 ԾᘏڗᜢᗖຒޑᆒዴϷєӣ Ҟ 500 ጇཥᆪϩ๏ 30 ঁԖङඳޑࣴز ғຑǴεऊԖ 58%ޑᜢᗖຒёௗڙǶ ૽ግኳࠠ܌ᘏڗрޑᜢᗖຒǴӧ 20 ঁᜢᗖ ຒޑєӣଯၲ 0.97ǴӢԜҁࣴز௦ڗጇЎകޑ 20 ঁᜢᗖຒǴٰीᆉځᜢᖄ܄Ƕ ٩Ᏽຒ༼ϐ໔ޑᜢᖄ܄Ǵڗрᜢᖄख़ଯܭ 0.5ޑޔௗϷ໔ௗᜢᖄຒǴӵ߄ 4Ǵ߄ 4 ύޑȸΌȹ ࢂғౢচǴکȸҡϯȹǵȸҡϯౢȹϷȸε ഌҡϯȹԖޔௗᜢ߯ǴԶȸҡϯȹຒΞჹᔈډ ȸᇸݨှቷȹǵȸၗीฝȹ ǵȸԃౢૈȹǴ ΨёаᇥࢂȸΌȹϷȸᇸݨှቷȹӧύЎཥ ᆪύԖ໔ௗᜢ߯Ƕ ߄ 4 ޔௗϷ໔ௗᜢᖄຒ ᜢᖄຒ ख़ ၌ ຒ༼ ޔௗ ໔ௗ ޔ ௗ ໔ௗ Ό ҡϯ ᇸ ݨ ှ ቷ 0.89 0.99 Ό ҡϯ ၗीฝ 0.89 0.73 Ό ҡϯ ԃౢૈ 0.89 0.73 Ό ҡ ϯ ౢ ᖄӝ௦ᖼ 0.75 0.89 Ό ε ഌ ҡ ϯ ٿ ۞ ҡ ϯ 0.63 0.75 Γ҇ჾϲ ॶႣය Ѧ ༊ Ӹۭ ᒲࢬΕ 0.66 0.72 Γ҇ჾສ ී Ѡ ༟ ჱݢΟය ݤ ୯ Ѓ ᎿሌՉ 1.00 1.00 Γ҇ჾສ ී Ѡ ༟ ჱݢΟය ᖄສਢ 1.00 0.87 Γ҇ჾສ ී Ѡ ༟ ჱݢΟය ᑼၗ 1.00 0.57 Γ҇ჾສ ී ύ ၗ ሌՉ Ѡᑼၗ 0.67 0.67 Γ҇ჾສ ී ύ ၗ ሌՉ ֎ ԏ Ѧ ༊Ӹී 0.67 0.67 Γ҇ჾສ ී ύ ၗ ሌՉ Ѧ ༊ Ӹ ීྗഢߎ 0.67 0.54 Γ҇ჾສ ී ύ ၗ ሌՉ ᖄສਢ 0.67 0.59 ύႝߞ ೯ ૻ ޣ ၰ ၡ ٬ Ҕ 0.58 0.67 ύႝߞ ଯ ೲ Ϧ ၡ ᇻܿႝη 0.57 0.75 ύႝߞ ଯ ೲ Ϧ ၡ ႝ η ԏ س 0.57 0.75 ѠՋ Ѡ ༟ ი ݅ ᚆ ཥᑫ 0.76 0.59 ѠՋ Ѡ ༟ ი ཥВ៓ 0.76 0.55 ѠՋ Ѡ ༟ ი က ᑜ ࡹ۬ 0.76 0.77 ѠՋ Ѡ ༟ ი Ꭶ ғ Ў ϯ 0.76 0.78 ѠՋ ྡྷᒳቷ ᄆ ϯ ε ࠤ ੇӦ 0.55 0.83 Ѡ༟ი ཥВ៓ ᒳ៓ቷ 0.55 0.96
5.
่ፕ
ҁࣴزޑჴբ่݀ǴளޕаΠ่ፕǺ (1) ύЎཥᆪЎകϣόߏǴᜢᗖຒޑቻ όܴᡉǴၨόܰᘏڗрٰǴӧҁࣴز ᗨฅԖၨଯޑєӣǴࠅࢂեᆒዴǶ (2) ҁࣴز܌ᘏڗрٰޑᗨฅόࢂ಄ӝΓ πۓကᜢᗖຒǴՠጇЎകԿϿԖϤԋ ޑຒ༼ёа٬ҔޣௗڙǶ(3) ҁࣴزаຒ༼ϐ໔ӅӕрޑᓎǴी ᆉຒ༼ᜢᖄ܄Ǵᡣ٬Ҕޣӧอਔ໔ջё ᘤំډύЎཥᆪޑᙁܰЇϷᜢᖄǶؒ Ԗک၌ຒ༼ӅӕрӧЎകύǴՠࢂ ࠅک၌ຒ༼ޑޔௗᜢᖄຒӅӕрӧ ЎകύǴΨёаբࣁ٬Ҕޣ၌ޑୖԵǶ
6.
҂ٰࣴزБӛ
ҁࣴز܌ᘏڗрޑᜢᗖຒϷᜢᖄຒǴ҂ٰᝩ ុࣴزӵՖᔈҔӧཥᆪЎҹᔠޑфૈǶࡌҥрཥ ᆪЎҹᜢᗖຒ༼ЇǴаკҢᡉҢόӕᜢᗖຒϐ໔ ޑᜢ߯Ǵڐշ٬Ҕޣזೲ၌܌ሡाޑཥᆪЎҹǶୖԵЎ
[1] Aas, K., Eikvil, L.: Text Categorisation: A Survey. Norwegian Computing Center, Oslo 1999 .
[2] Dutta, S. and Shekhar, S., Bond rating: A
non-conservative application of neural networks, IEEE International Conference on Neural Networks-San Diego, Vol.2 , pp443-450, 1988. [3] Krulwich, B., and Burkey, C. , Learning user
information interests through the extraction of semantically significant phrases. In M. Hearst and H. Hirsh, editors, AAAI 1996 Spring Symposium on Machine Learning in Information Access. California: AAAI Press.
[4] Muñoz, A., Compound key word generation
from document databases using a hierarchical clustering ART model. Intelligent Data Analysis, 1 (1), Amsterdam: Elsevier.1996.
[5] Ma, Wei-Yun and Keh-Jiann Chen, Introduction to CKIP Chinese Word Segmentation System for the First International Chinese Word Segmentation Bakeoff, Proceedings of ACL, Second SIGHAN Workshop on Chinese Language Processing, pp168-171, 2003.
[6] Odom, M, Sharda, R., Aneural network model for bankruptcy prediction,IEEE INNS IJCNN,Vol.2,PP.163-168,1990.
[7] Salton,Gerard , Automatic text processing:the transformation, analysis, and retrieval of information by computer,Addison-wesley publishing Company, Inc,1989.
[8] Steier, A. M., and Belew, R. K. , Exporting phrases: A statistical analysis of topical language. In R. Casey and B. Croft, editors, Second Symposium on Document Analysis and Information Retrieval, pp. 179-190, 1993.
[9] Turney, P.D., Extraction of Keyphrases from Text: Evaluation of Four Algorithms. National Research Council, Institute for Information Technology, Technical Report ERB-1051,1997.
[10] Turney, P.D., Learning to Extract Keyphrases from Text. National Research Council, Institute for Information Technology, Technical Report ERB-1057,1999.
[11] Turney, P.D. Learning algorithms for keyphrase extraction. Information Retrieval, 2, pp.303-336, 2000.
[12] Witten, I.H., Paynter, G.W., Frank, E., Gutwin, C. and Nevill-Manning, C.G., KEA: Practical automatic keyphrase extraction. Proceedings of Digital Libraries 99 (DL'99), pp. 254-256. ACM Press,1999. [13] මϡᡉǴᜢᗖຒԾᘏڗמೌᆶ࣬ᜢຒӣ㎸Ǵ ύ୯კਜᓔᏢൔǴ1997 ԃǴ12 ДǴಃϖ ΜΐයǴ। 59-64Ƕ [14] මϡᡉǴᜢᗖຒԾᘏڗמೌϐǴύ୯კ ਜᓔᏢૻǴ1997 ԃǴ9 ДǴಃ 106 යǴ । 26-29Ƕ