基于E-OEM的SWF半結(jié)構(gòu)化模型建立
發(fā)布時(shí)間:2019-05-21 21:31
【摘要】:為了揭示SWF文件格式的隱含屬性,基于對(duì)象交換模型(OEM)的構(gòu)建方式,提出了一種增強(qiáng)半結(jié)構(gòu)化模型E-OEM,可對(duì)SWF文件格式進(jìn)行描述和存儲(chǔ).采用OEM進(jìn)行SWF文件格式的描述;對(duì)OEM描述模型進(jìn)行改進(jìn),采用頭尾分類、文件體聚類的思想將相同類別標(biāo)簽作為一類;通過引入Huffman編碼,實(shí)現(xiàn)E-OEM具有描述和存儲(chǔ)的功能.隨機(jī)選擇百例不同源文件進(jìn)行E-OEM建模,仿真實(shí)驗(yàn)結(jié)果表明,所提模型不僅可以將隱含屬性顯性表示,同時(shí)提高了具有高標(biāo)簽重復(fù)率的文件存儲(chǔ)效率,證實(shí)了模型的有效性.
[Abstract]:In order to reveal the implicit attributes of SWF file format, based on the construction of object exchange model (OEM), an enhanced semi-structured model E / OEM is proposed, which can describe and store the SWF file format. OEM is used to describe SWF file format, OEM description model is improved, and the same category label is regarded as a class by using the idea of head and tail classification and file body clustering. By introducing Huffman coding, E-OEM has the function of description and storage. 100 different source files are randomly selected for E-OEM modeling. The simulation results show that the proposed model can not only express implicit attributes explicitly, but also improve the file storage efficiency with high label repetition rate, which verifies the effectiveness of the model.
【作者單位】: 北京郵電大學(xué)數(shù)字媒體與藝術(shù)設(shè)計(jì)學(xué)院;新疆師范高等專科學(xué)校新疆教育學(xué)院;
【基金】:國家科技支撐計(jì)劃項(xiàng)目(2014BAH13F02)
【分類號(hào)】:TP311.13
,
本文編號(hào):2482414
[Abstract]:In order to reveal the implicit attributes of SWF file format, based on the construction of object exchange model (OEM), an enhanced semi-structured model E / OEM is proposed, which can describe and store the SWF file format. OEM is used to describe SWF file format, OEM description model is improved, and the same category label is regarded as a class by using the idea of head and tail classification and file body clustering. By introducing Huffman coding, E-OEM has the function of description and storage. 100 different source files are randomly selected for E-OEM modeling. The simulation results show that the proposed model can not only express implicit attributes explicitly, but also improve the file storage efficiency with high label repetition rate, which verifies the effectiveness of the model.
【作者單位】: 北京郵電大學(xué)數(shù)字媒體與藝術(shù)設(shè)計(jì)學(xué)院;新疆師范高等專科學(xué)校新疆教育學(xué)院;
【基金】:國家科技支撐計(jì)劃項(xiàng)目(2014BAH13F02)
【分類號(hào)】:TP311.13
,
本文編號(hào):2482414
本文鏈接:http://www.sikaile.net/kejilunwen/ruanjiangongchenglunwen/2482414.html
最近更新
教材專著