@techreport{oai:ipsj.ixsq.nii.ac.jp:00194717,
 author = {本郷, 節之 and 大加瀬, 稔 and 寺田, 雅之 and 鈴木, 昭弘 and 稲垣, 潤},
 issue = {36},
 month = {Feb},
 note = {Privelet 法は，差分プライバシ基準に準拠しつつ，部分和精度にも優れており，プライバシが保護されたデータのスケーラブルな活用を可能にする．しかし 「非負制約の逸脱」 や 「疎データの密度急増」 という問題は回避できない．けれども，この Privelet 法に非負精緻化処理を組み込むと，高い部分和精度を維持しつつ，これらふたつの問題への対処も可能となる．この手法の場合，非負精緻化を伴う逆 Wavelet 変換 （Top-down 精緻化） 部分に枝刈り処理を導入することで演算を効率化することができる．筆者らは以前，Top-down 精緻化の性質に着目した枝刈り実装法 （水平型） を提案した．本報告では，先の提案とは異なる実装法 （垂直型） を新たに提案する．さらに，先に提案した実装法との間での，演算効率化効果の比較評価も試みる．, Privelet is a data publishing technique that ensure ε-differential privacy while providing accurate answers for range-count queries. This technique is suitable for scalable utilization of privacy-preserved data. However, it has two problems which are “deviation from the non-negative constraint” and “abruptly increase of data-density”. Our non-negative refinement solves these two problems without losing the accuracy of the partial summation. In this method, it is possible to improve the efficiency of calculation by introducing pruning processing in the inverse Wavelet transform with nonnegative refinement - the top-down refinement. We have proposed a pruning implementation method - the horizontal type - focused on characteristics of the top-down refinement. In this report, we propose a new implementation method - the vertical type - different from the previous proposal. Additionally, we will try to compare and evaluate the efficiency improvement effect with the previously proposed implementation method.},
 title = {非負精緻化を伴うPrivelet法演算効率化の試み},
 year = {2019}
}