資源描述:
《關(guān)鍵詞拍賣的選詞與出價(jià)策略研究-論文.pdf》由會(huì)員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在行業(yè)資料-天天文庫。
1、第4l卷第7期計(jì)算機(jī)工程2015年7月VO1.41NO.7ComputerEngineeringJuly2015·開發(fā)研究與工程應(yīng)用·文章編號(hào):1000-3428(2015)07-0310-07文獻(xiàn)標(biāo)識(shí)碼:A中圖分類號(hào):TP391關(guān)鍵詞拍賣的選詞與出價(jià)策略研究吳紀(jì)蕓,陳志德一,汪磊一.王孟,(福建師范大學(xué)a.?dāng)?shù)學(xué)與計(jì)算機(jī)科學(xué)學(xué)院;b.網(wǎng)絡(luò)安全與密碼技術(shù)福建省重點(diǎn)實(shí)驗(yàn)室,福州350000)摘要:在關(guān)鍵詞拍賣中,每個(gè)廣告主都有成千上萬的關(guān)鍵詞可選,為了在預(yù)算限制條件下給選擇的關(guān)鍵詞設(shè)置合理的投標(biāo)價(jià)格,提出一種新的基于廣告主的拍賣策略,包括選詞策略和出價(jià)策略。在選詞策
2、略中,提出基于詞頻一反轉(zhuǎn)文件頻率算法的關(guān)鍵詞關(guān)聯(lián)度計(jì)算方法,通過該方法選出的關(guān)鍵詞不僅能提高網(wǎng)站的關(guān)聯(lián)度,增加轉(zhuǎn)化率,還能避免因使用過度普遍的關(guān)鍵詞而增加競(jìng)爭(zhēng)成本。在出價(jià)策略中,運(yùn)用改進(jìn)的粒子群優(yōu)化算法,在若干約束條件限制下對(duì)每個(gè)關(guān)鍵詞的出價(jià)做適當(dāng)調(diào)整,以增加廣告主所獲利潤。實(shí)驗(yàn)結(jié)果表明,采用拍賣策略選出的關(guān)鍵詞組可增加網(wǎng)站的轉(zhuǎn)化率,降低競(jìng)爭(zhēng)成本,所獲得的利潤比傳統(tǒng)人為投標(biāo)所獲得的利潤高,并且在初期和中期呈現(xiàn)持續(xù)上升趨勢(shì),后期趨于穩(wěn)定。關(guān)鍵詞:關(guān)鍵詞拍賣;選詞策略;關(guān)聯(lián)度;出價(jià)策略;詞頻一反轉(zhuǎn)文件頻率算法;粒子群優(yōu)化中文引用格式:吳紀(jì)蕓,陳志德,汪磊,等.關(guān)鍵詞
3、拍賣的選詞與出價(jià)策略研究[J].計(jì)算機(jī)工程,2015,41(7):310.316.英文引用格式:WuJiyun,ChenZhide,WangLei,eta1.StudyonWordSelectionandBiddingStrategiesofKeywordAuction[J].ComputerEngineering,2015,41(7):310—316.StudyonWordSelectionandBiddingStrategiesofKeywordAuctionWUJiyun一,CHENZhide一,WANGLei,WANGMeng'(a.CollegeofM
4、athematicsandComputerScience;b.FujianProvinceKeyLaboratoryofNetworkSecurityandCryptography,F(xiàn)ujianNormalUniversity,F(xiàn)uzhou350000,China)【Abstract】Therearemillionsofavailablekeywordsforeachadvertiserinkeywordsauction.Howtosetareasonablebidpricefortheselectedkeywordsunderalimitedcondition
5、,suchasbudget,becomesthemostdifficultworkfortheadvertiser.Asitishardformanyadvertiserstoselectkeywordsandsettheprice,anovelmodelofauctionstrategybasedonadvertisersisproposedfortheseproblems.Thisauctionstrategyincludesthekeywordsselectionstrategyandthebiddingstrategy.Thekeywordsselect
6、ionstrategypresentsacalculationmethodforkeywordscorrelationwhichisbasedontheTermFrequency—InverseDocumentFrequency(TFIDF)algorithm.Thekeywords,selectedthroughthismethod,notonlyimprovethecorrelationwiththepromotedwebsite,increasetheconversionrate,butalsoavoidincreasingthecompetitionco
7、stduetotheoveruseofcommonkeywords.ThebiddingstrategyusesanimprovedParticleSwarmOptimization(PSO)algorithmtoproperlya0justthebidsofeachkeywordundersomeconstraintconditionsSOastoincreasetheprofitsofadvertisers.Experimentalresultsshowthatkeywords,selectedthroughtheauctionstrategy,increa
8、setheconvers