【摘要】：隨著網(wǎng)絡(luò)視頻爆發(fā)式增長(zhǎng),在線視頻服務(wù)資源面臨著嚴(yán)重過(guò)載,準(zhǔn)確預(yù)測(cè)視頻播放量對(duì)供應(yīng)商而言越來(lái)越重要。論文通過(guò)對(duì)實(shí)際在線視頻服務(wù)系統(tǒng)所采集的海量數(shù)據(jù)研究,針對(duì)視頻放映的不同時(shí)間段,分為上映前的精準(zhǔn)預(yù)測(cè)和上映后的同步預(yù)測(cè)二個(gè)階段:1)視頻上映前,針對(duì)傳統(tǒng)預(yù)測(cè)模型分類(lèi)和預(yù)測(cè)效果不佳、規(guī)則化較多和缺乏實(shí)際應(yīng)用價(jià)值等問(wèn)題,提出一種基于深度信念網(wǎng)絡(luò)(Deep Belief Networks,DBNs)的視頻播放量預(yù)測(cè)方法。首先,結(jié)合社交網(wǎng)絡(luò)的關(guān)注度和視頻關(guān)鍵詞的搜索熱度,對(duì)影響因子進(jìn)行建模和量化處理;其次,根據(jù)輸入和輸出變量確定DBNs各層網(wǎng)絡(luò)的結(jié)構(gòu),優(yōu)化網(wǎng)絡(luò)參數(shù)和預(yù)測(cè)模型;最后,利用在線視頻服務(wù)商的數(shù)據(jù)對(duì)深度信念網(wǎng)絡(luò)進(jìn)行訓(xùn)練,并多次交叉實(shí)驗(yàn)對(duì)比分析,結(jié)果表明基于DBNs方法在視頻播放量預(yù)測(cè)準(zhǔn)確率上有較大提升,有效實(shí)現(xiàn)了視頻播放量的早期預(yù)測(cè)。2)視頻上映后,通過(guò)對(duì)在線視頻早期播放量時(shí)序的統(tǒng)計(jì)分析,提出一種基于ARMA模型的視頻播放量預(yù)測(cè)方法,同步預(yù)測(cè)視頻未來(lái)某天的播放量。根據(jù)視頻播放量時(shí)序特征的差異性選擇不同的預(yù)測(cè)模型,模型在對(duì)非平穩(wěn)的國(guó)內(nèi)視頻和季節(jié)性明顯的國(guó)外視頻日播放量的同步預(yù)測(cè)獲得了較高精確度,相比傳統(tǒng)的移動(dòng)平均法、指數(shù)平滑法和最小二乘法的預(yù)測(cè)方法獲得了明顯的提升,具有實(shí)際的參考價(jià)值。通過(guò)對(duì)深度信念網(wǎng)絡(luò)和時(shí)間序列模型的研究,本文實(shí)現(xiàn)了在不同時(shí)間階段對(duì)視頻播放量進(jìn)行及時(shí)、持續(xù)、準(zhǔn)確的預(yù)測(cè),既能為視頻上映前的投資、評(píng)估提供較全面可靠的參考決策;又能夠得到上映后未來(lái)時(shí)間點(diǎn)精確的播放量波動(dòng)范圍,為設(shè)計(jì)合理的廣告投放、資源存儲(chǔ)和商業(yè)決策提供支持。
[Abstract]:With the explosive growth of online video, online video service resources are facing serious overload, so accurate prediction of video playback is becoming more and more important for suppliers. Based on the research of massive data collected by the actual online video service system, this paper is divided into two stages: accurate prediction before release and synchronous prediction after release according to the different time periods of video screening: 1) before the release of video, In order to solve the problems of poor classification and prediction effect of traditional prediction model, more regularity and lack of practical application value, a video playback prediction method based on deep belief network (Deep Belief Networks,DBNs) is proposed. Firstly, the influence factors are modeled and quantified according to the attention of social network and the search heat of video keywords. Secondly, the structure of each layer of DBNs network is determined according to the input and output variables, and the network parameters and prediction model are optimized. Finally, the data of online video service providers are used to train the deep belief network, and many cross experiments are compared and analyzed. The results show that the accuracy of video playback prediction based on DBNs method is greatly improved. The early prediction of video playback is effectively realized. 2) after video release, through the statistical analysis of the timing of online video early playback, a video playback prediction method based on ARMA model is proposed. Synchronously predict the amount of video to be played one day in the future. According to the difference of time series characteristics of video playback, different prediction models are selected. The model obtains higher accuracy in the synchronous prediction of non-stationary domestic video and seasonally obvious foreign video daily broadcast volume, compared with the traditional moving average method. The prediction methods of exponential smoothing method and least square method have been improved obviously and have practical reference value. Through the research of deep belief network and time series model, this paper realizes the timely, continuous and accurate prediction of video playback at different time stages, which can not only invest in video before release. The evaluation provides a more comprehensive and reliable reference decision; It can also get the accurate fluctuation range of broadcast volume at the future time point after release, and provide support for the design of reasonable advertising, resource storage and business decisions.
【學(xué)位授予單位】：深圳大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 袁楚;;手機(jī)在線視頻似近實(shí)遠(yuǎn)[J];互聯(lián)網(wǎng)天地;2009年04期

2 王道才;;不用工具 在線視頻輕松拿下[J];電腦愛(ài)好者(普及版);2009年04期

3 ;艾瑞咨詢(xún):2012年中國(guó)在線視頻行業(yè)六大盤(pán)點(diǎn)[J];電視技術(shù);2013年02期

4 鄧祖平;;在線視頻廣告監(jiān)管系統(tǒng)[J];廣播與電視技術(shù);2013年02期

5 丁佳琪;;中國(guó)在線視頻行業(yè)發(fā)展分析[J];傳媒;2014年06期

6 南湖秋水;;在線視頻體驗(yàn)影院模式[J];網(wǎng)友世界;2010年10期

7 loading;;從在線視頻用戶中消失[J];電腦迷;2004年10期

8 日出東方;;在線視頻 不能沒(méi)有你[J];電腦迷;2006年14期

9 惜朝;;流暢看電影,優(yōu)酷優(yōu)跳不(YouTube)都別卡[J];電腦愛(ài)好者;2009年02期

10 Aa醬;;向在線視頻提要求[J];電腦迷;2009年06期

相關(guān)會(huì)議論文 前1條

1 柴焱;李s，

本文編號(hào)：2473924