一個具有模糊查詢能力之物件導向資料庫查詢語言(EQLF)的設計與製作
全文
(2) .
(3) (EQLf) The Design and Implementation of an Object-Oriented Database Query Language (EQLf) for Fuzzy Querying : NSC87-2213-E009-001
(4) : .
(5) !"#$%&'( ) *+,-./0123 45 #67"89: 7;<=#$% 45(>?@ )A.2BCD EF 4589:7G./01"F#$DEGH IJK # 45"LM?@ NOPQRS 45 89:7 TD"FU.VW%XYZ[\89]^ D_2 ` a bc%deLMbf^gh 45 i.jklmZ[\89]^n89:7"opq rstuJvwxyzMa{|L}~%N Z[\89 N"o n( }RSLM f^gh 4589:7 "F U.jkZ[\89 TD"o3LZ[\ 89XYJ}f^ghAn o ¡¢* +nc wdezM b#0. £i/01¤¥¦nlm89:72§¨zM89 : 7 ©ª B EQLf (An Easy-to-Use Query Language with Fuzzy Capability for Object-Oriented Databases)2 Abstract With the popularity of applied information technology, the user interface that acts the role for communications between the users and the computers has been designed to be more and more friendly. For users who want to access a database, the query language fits as the most important user interface. To make the query language more user-friendly, an important issue arose is to extend the query language with the functionality to accept and explain imprecise queries. The objective of this project is to design and implement an object-oriented database query language for fuzzy querying we achieve this objective through the following steps; 1) Analyze and discuss researches on processing of imprecise queries and compare the advantages and disadvantages among them; 2) Extend an existing query language, namely EQL, to incorporate the capabilities of expressing imprecise predicates and making imprecise queries; 3) Implement the complete query language to validate its excellence in being both. user-friendly and easy to learn by adopting object-oriented engineering technology and windows programming paradigm.We call this query language EQLf (An Easy-to-Use Query Language with Fuzzy Capability for Object-Oriented Databases). . . b 45 «¬"89:7;<=#$% 4 5() ®¯"°Q±² N³ 45´µb¶· f^gh 45"¸XY¹@ 4 JQ #º»¼½"¾¿ZQÀy89ÁÂÃsÄÅ"PQ ÆvÇ ÈÉlb 45²i"#Ê 45ÆvVW ËÌÍÎ >Ì# Z[\8 92389Z [\1 "ÏÐÑ N "oÒÓebLÔÓÕ Ö×i"zÔØÙX YZ[\]^ J"Út~B30Û3ÜJ v2 L}3Z[\ÝÞßàáâãäßåæçèßéãêãëìí. XYv. î 3 Z[ \PQ Zï ðM yñ LM3 Z[\]^ òó ô"6Qï MyñLM by ñ õª"ö LMyñ UyñG÷ ø¿Z[ \]^"ù¿yñ õªPú Uyñû "üùP Uyñý23 Z[\] ^þ 3XY JvbÓeiQ I "¾ BU Q byñ õi" °89 Á Â%§ oZ 2 LM PQMl m:¤
(6) QLM yñ1 "¾¿l m]^ 1PúW 2 }Û3Z[\Ýêëàâßåæçèßéãêãëìí. XYv. î Û3Z[\Ýêëàâßåæçèßéãêãëìí"P QïðMyñLMV òóô"w¿yñ3Z [\]^ 2ØÙ# Jv Z" Ú~BÜ "LBt \! Ý"èãêç#ßâêíB " L Btlm! Ý$%%&#ßâêíB 2 'ti Ú t3zÔJ v()U "§t*++w~ zÔJv ,"M-Ñ./ - ">0 ÁÂöL 1 Relative. Absolute Interpretation.
(7) Interpretation. Deduce2. Crisp Sets Based ARES Vague **. FQL, SQLf ***. Extra Knowledge. *. Integration. **. *. *. ***. Intuition. *. **. ***. ****. Implementation Effort. *. **. **. ***. L1Z[\89. **. Fuzzy Sets Based. M",PQ[@|i o M>÷ø f^"u * } r ~ n( ô",PQ[@|( u f ^2 duJ $%%&pèß5ãéáâßê ÚtQlm] ^0º»]^ "l m]^ TUÚtb lm] ^}µlm]^} lm]^tllmN2b á ô(ÚtØÙJ 89 ÁÂ"3 C f Ý~í Y\6Ê2 µ C f (a ) =. Sup. o∈C and o.Ai = a. XYJv . ❶2âèá3ìë4àß56ß| ¿Jvb±² 45i Ñ7Ñ89:
(8) ;DXYUD Z[\ 892 ❷åìâß6èáâãëì| ¿Jv3Z[\]^< x=1"QüZ [\]^¤Út<0XY2 ❸åìâãâãëì| ¿Jv890 ÁÂQü%§ 2 ❹åæçàßæßìâáâãëì>>ëèâ| ¿JvbÓei ?I2 `@ØÙtiAB ÁÂ"CDbx=10 1¤¦ JvEtlm! B XYJ v"wcdeLM.lm89D_nf^gh 4589:7 >2. µFuzzy Pr edicates (o ) . o. iBJÝ~í" c " f^"p/Bj f^ p b1 Bj i no" a ∈C . Ai " C f l mcÝ$%%&"àáêêí"¸f^31 Bj no 0 ÁÂÚDúwÍ?ô" Sup Ý#çèßæßí IKP Q3zÔ?ôþ >iô 2Vuw§ LÔ N "o ØÙJ Ý~ íw \M f^ ( 2 $~í. Í 4. Student age = "old" bzM"lm ]^B "¾¿§¨ Mf^ ` ("o7t\"P ÚtÊf^ (B µStudent (o) = µold (o.age) 2 (
(9) ¸#$´²$tJ K2 f. .
(10) > Q¸ :7RS6w"FG > HlmÝ$%%&í nI"zoZQ > QLMJKZLM}:vlm :7"UN@< Q"OC P TD:"> QDLM jkZ[\]^"o7t2 tu§R3 > 3 S RSTU7tV " OC3:v XY:"G?@ Q3WZ lm8 9"§,ABXXYCU :I",PQJK3 Y2 !"# b Zþ ]^[DQ±² º»Ç\Ý]ëëàßáì 2çèßêêãëìí"¾¿§Út^¸_`lm]^"6ay î lmCD2zQ>b Llm89"ÁaB|.
(11) b " QwjkÜd( "#. B“ ”"zM w³Á Ü1" ÝßãäEqëãìí2> R3 n fuzzy relation eLRS"¨Pw B“ ”"l m "¾¿89 ÁaB| Class Name C1. Class Name C2. Attributes A1. Attributes A1. . . .. Attributes Ai . . .. Attributes An Predicate fc1. . . .. fuzzy relation. Attributes Aj . . .. Attributes An Predicate fc2. . Class Name C Attributes A1 . . .. Attributes Ai . . .. . Attributes An Fuzzy Predicates. b > LMc¤¸LMdeÝfßèâß2åéëìí 7t"gC¿c01 ªht3¿ c@u< Zþ]^21ijk lm “ ” @noÝpèëqßéâãëìí 12¿ÁaH IKPQ3c " f^"ØÙ]^\U ( "o¨Unor1 Bj i2 > M89bsti¤Út¸#$uJ>0 ÁÂ@övwÍ"uJx ×Byozuzo-u!{"o. lm Üd~cBLMlmCD>%%& èßàáâãëì ÚtBv l \N´lmN" ö|“” "“” ÑlmN" “ ” lmN2bzM89 Á "3 C f Ýáíô"(ÚtØÙJÝ¡í Y7t\6Ê2. ≈. µC (a, b)= f. Sup. oC1∈C1andoC1.Ai=aand oC2∈C2andoC2.Aj =b. min( µfc1 (oC1),µfc2 (oC2),µfuzzyrelation(oC1.Ai ,oC2.Aj )). JÝ¡í oC1 c "~ f^" oC2 c "¡ f^" C f lmcÝ$%%&"àáêêí2LôÊ §¢I Q"c "~ "¡"Bj Bk"6B.
(12) NO1"tlmNB“”"lm Útlm CD7tH£"¤@ Class Name C Attributes A1 . . .. Class Name C Attributes A1. Class Name C. Attributes Ai .. Attributes Ai .. Attributes An Predicate fc2. Attributes An fc1 and fc2. Attributes A1. . . .. Attributes Ai .. =. . .. . . .. . .. . .. Attributes An Predicate fc1. JÝÁí o "~ nf^"p/Bj f^ p b 1 Bj i no" a ∈ C1. Ai " C f lmcÝ$%%&"àáêêí" ori Ò o xÉn "¡ f^ n Ò o xÉnf^M 2 ¡ÀlmNlm¹ºB31. 2 n. µCf (a) = Sup min(µfc1(o),µfuzzyquantifier(1/n∑µfc2(o.ori )) o∈C1ando.Ai =a. ¿§P
(13) É¥89L¦1Ýßè&ãäáàßìéßí" ,PQVf^ §¸¨©l m \ ( @ª ©lm89\ ("z«JKY; Q 2 ¬§w®¡¨© YB| µC (a) = f. Sup. oi∈Candoi. Ai=a and oj∈Candoj .Ai=a. min(µfc1 (oi ),µfc2 (oj ),µ= (oi .Ai ,oj .Ai )) . i =1. JÝÂíBY" o "~ nf^"p/Bj f^ p b1 Bj no" a ∈ C1. Ai " C f lm cÝ$%%&"àáêêí" ori Ò o xÉn "¡ f^ n Ò o xÉnf^M2 ÁÀlm]^ >é¡ lm¹ºBÛ31 2Y JKB| µ C ( a) = f. Sup. o∈C 1and o. Ai = a. min( µ fc1 (o ), n. µ fuzzy quantifier (∑ µ fuzzy relation (o.Ai , o.o ri. A j )) i =1. 6ª© µC (a ) = f. Sup. YB|. o∈ C and o . A i = a. min( µ fc 1 (o ), µ. ( o )) fc 2. §Út¶zÜMYQL« "¾¿÷ø89L ¦12Vuw|M1 $¡íH¯°±ZÑ Professor. ²³ 4. ~~. µC (a ) =. Salary age = "young". f. f. Sup min(µyoung(oi .age), µyoung(oj.age),µ≈ (oi.salary,oj.salary)). oi,oj∈Professor.
(14)
(15) b " Ñ ô 1 ݵàâãäáàßí 0 L 3 Ñ Ý¶ìßEâëEæáì&í N ·J J Q ¸b1 Ý2ãêâßìâãáàí¹º7t"öÂ#$É¥ Q»¬ 1ݼìãäßèêáàí¹º"P
(16) b89 cªhiuJ» ¬¹º “ ∀ ”2¾¿§3 n¸b10»¬1 ¹ºeLRS"½Ðb89#lm¹º"89 Áa öu| reference. Class Name C1 Attributes A1 Attributes Ai .. fuzzy quantiferClass Name C2 Attributes A1 . . .. . . .. fuzzy relation. . .. Attributes An Predicate fc1. Attributes Aj . . .. Attributes An Predicate fc2. ØÙ#$3¿89 #J"( \Út~B ¾¿×2 ~ÀlmNlm¹ºQÛ31 2 YJKB|. µC (a) = f. Sup. o∈C1and o. Ai =a. Sup. o∈C1 and o. Ai = a. 2Y. min(µ fc1 (o), n. bzM" QÜMlmCD" JQ3 ²³ ¯°Slm " NB±ZÑ"¾¿ §ÚtØÙ´Ý¡í"Ú\ 0 f^(1 µC (oi ,oj ) =. o. cÝ$%%&"àáêêí" ori Ò xÉn "¡ f^! o.ori. Aj f^ o.ori b1 Bk i no" n Ò o xÉnf^M2 ÂÀlm]^ >é¡ lm¹ºB31 JKB|. Professor. Salary age = "young". iBJÝÃí" o "~ nf^"p/Bj f^ p b1 Bj i no" a ∈ C1. Ai " C f lm. n. min(µ fc1 (o), µ fuzzyquantifier (∑ µfc 2 (o.ori )) i=1. µ fuzzyquantifier (1 / n ∑ µ fuzzyrelation (o. Ai , o.o ri. A j )) i =1. iBJÝÄí" o "~ nf^"p/Bj f^ p b1 Bj i no" a ∈ C1. Ai " C f lm cÝ$%%&"àáêêí" ori Ò o xÉn "¡ f^! o.ori. Aj f^ o.ori b1 Bk i no" n Ò o xÉnf^M2 Ã.lm]^ >é¡ lmNlm¹ºB31 2 YJKB| µ C ( a) = f. Sup min(µ fc1( o),. o∈C 1ando.Ai=a. n. µ fuzzy quantifier(. ∑min(µ. i= 1. fuzzy relation. (o.Ai , o.ori .A j ),µ fc2 (o.ori )). n. ∑ µ (o.o ) i= 1. fc 2. )). ri. JÝÅí "~ nf^"p/Bj f^ p b1 Bj i no" a ∈ C1. Ai " C f lmcÝ$%%&"àáêêí" ori Ò xÉn "¡ f^! o.ori. Aj f^ o.ori b 1 Bk i no" n Ò o xÉnf^M2VuwQ LÔN 2 $ÁízM 4"]^B TU ÆÇQÈû û 2. o o. Person. friend. " most". Person. height = "tall". zM89 lm¹ºB TU"¿89 Áa É89"tØÙ´ÝÂí"f^ (B1.
(17) n. µPerson (o ) = µmost(1/ n ∑ µtall (o. friend i .height )) 2 û0 f. i=1. TU. (¹¸#$7tJK2.
(18) ¨i+* Ê lm8 9Ë Ì! eÁ 0P yBz 89"89ÁaB ∪. Class Name C1 Attributes A1. . . .. . . .. Attributes Ai .. Attributes An Predicates fc1. Attributes An Predicates fc2. . .. . ..
(19) . tu§bV > ²WTU @TD| ~> #$*+Ý>¼êßèåìâßè>áéßí1zTUÖ ×%#$)"@* TD| 45 ³Á%X O}89 }lm:I }89Á ·2 ¡lmØÝ$%%&µáìá6ßèí1¿ØÖ×Ù¸} #$ lm:I"oÚstØ89lm: I *+2 Á89ØÝßè&pèëéßêêëèí1¿ØX# $u< 89"oØÙ > cÛJM89 Y"oÜstØY 2. Class Name C2 Attributes A1. Attributes Ai .. . . §ÚtËlm! Ì!ÍÎwXY¿8 9"b > Qt æá2 wXYÌ!Ç\"¾¿Y JKB1. ÂstØÝ2ßéâãëìµáìá6ßèí1ØÙ89ØÝ 0 ÁÂ"¨ 45 f^þ"o\(" 0¨Á±Þï#$*+2 User. EQLf User Interface. semanti c offuzzyt erm. µ C (a ) = max( µ fc1 (a ), µ fc 2 (a )) . parsing. f. JÝÏí $ÂíH. ²³0Ð. ∪. Professor. identi ficati on FuzzyManager. Ѫ2 membership function of fuzzy term. Person. name. Query Processor notifi cation mat ching funct ion. Executi onManager. name. age = "young". ØÙ´ÝÏí"f^. salary = "high". Obj ectPro/ODB. (B:. eL1> ²ßae. µC (a) = max(µyoung (a), µhigh (a )) ƌ. FRMg. f. $%&'(.
(20) . ,. -. . 4. /.
(21)
(22). . . &')*. . 1 Ofx!Rvfsz!3. R Ofx!Rvfsz!2. ¨i+* Ê lm8 9Ë ±! eÁ 0P yBz 89"89ÁaB Class Name C1 Attributes A1. Class Name C2 Attributes A1. Attributes Ai .. Attributes Ai .. Attributes An Predicates fc1. Attributes An Predicates fc2. +,-. . §ÚtËlm! "!0Ò! ÍÎwXY¿ 89"b > Qt æãì wXY!Ç\"t{Ólm ÔÒÇ\wXYÔÒ"¾¿YJKB1 µC ( a ) = min( µ fc 1 ( a ),1 − µ fc 2 (a )) f. JÝÕí $ÃíаZQH²³. . Ѫ2. Professor. Person name. name. salary = "high". age = "young". ØÙ´ÝÕí"(B1 µC (a) = min(µhigh (a ),1− µ young (a)) f. . e1> *+:à. . .. . .. ∪ . . . .. . . .. . .,-. `ábc%deLMbf^gh 45. jkZ[\89]^n89:7"§âCLÔ Z[\]^ Jv"o~()U "¸ lm89>` @ã@
(23) ÷ø#$ 1"o 3Wlm89äDÚ XY"tlm! B XYJbzÜ @ãi"ä.>¦ "¾¿§¨lm! blm89 XYi2 b:7cJ+"§C Útlmå T U"tZoæ PwçkBi"Ëlm! NÍÎwJKWlm89 Y"§CDC> Ì# lm! èeÇ\wc§ Y"p q89 ÁÂGD #$ ¹ã2 ai§Ï3 > Úlmå TU7tRS"l m89 >¦å¨Q§uLB é2. !"#.
(24)
相關文件
The objective of this study is to investigate the current status of the elementary school teachers who promote Gratitude Education by using Gratified Dream Savings Notebook
The objective of this study is to analyze the population and employment of Taichung metropolitan area by economic-based analysis to provide for government
The main objective of this article is to investigate market concentration ratio and performance influencing factors analysis of Taiwan international tourism hotel industry.. We use
This research project is to analyze and research a specific laser configuration to obtain complete knowledge、technologies and experiences to transfer to the project initiator for
The main purpose of this paper is using Java language with object-oriented and cross platform characteristics and Macromedia Dreamweaver MX to establish a JSP web site with
The main objective of this system is to design a virtual reality learning system for operation practice of total station instrument, and to make learning this skill easier.. Students
The objective is to evaluate the impact of personalities balance in a project management team on the team’s performance.. To verify the effectiveness of this model, two
The purpose of this study is to analyze the status of the emerging fraudulent crime and to conduct a survey research through empirical questionnaires, based on