英语翻译2 Related WorkVarious methods have been proposed for tim

英语翻译2 Related WorkVarious methods have been proposed for timeseries searching.In [l],an indexing scheme calledF-index is suggested to handle data sequences andquery sequences of the same length.Firstly,eachdata sequence is transformed by n-point DiscreteFourier Transform.The first fc coefficients arekept and regarded as a fc-dimensional point.Thefeature points are then indexed by an R*-tree[16].For a range query,the query sequence isfirst mapped to a point in the fc-dimensionalspace similarly.Then,the R*-tree is searchedand all feature points that are within the errordistance from the query sequence are retrieved.This method guarantees no false dismissal,but itmay cause false alarms.Thus,the original Idatasequences corresponding to the points retrilevedhave to be checked against the query sequence.The results in [l] are further generalized in [2]and the ST-index is proposed to handle data sequencesof different lengths.A sliding window withlength n is placed over the data sequences.Thesubsequence within each window is transformedby n-point Discrete Fourier Transform.After alldata subsequences are transformed,a trail will beformed.The trail is divided into sub-trails,whichare then represented by minimum bounding rectangles(MBR) o f an R*-tree.For range query,al1MBR that intersect the query region will be retrieved.This method also guarantees no false dismissal,yet false alarms are still possible,and so theoriginal data sequences have to be checked againstthe query sequence,too.The methods proposedin [l,21 are very elegant.However,they use EuclideanDistance for sequence similarity withoutconsidering any transformation.As shown by theexample in Section 1,it is better to consider sequencesimilarity with scaling and shifting in someapplications such as stock analysis.The definition of similarity used in this paper issimilar to those proposed in [4].In [4],the authorsdevelop a general framework for similarity queries.The framework consists of a transformation rulelanguage T.An object A is said to be similar to anobject B if A can be transformed to B by a series oftransformations defined in T.Each transformationapplied has a cost and the total cost is used tomeasure the distance between A and B.In [5],the authors consider the case that T containsthe transformation of moving average andtime wrapping.They first show that the definitionof sequence similarity with moving averageand time wrapping has a wide range of real applications.Then,they illustrate by real stock datathat the transformations help to identify similarruns of stock price.They also propose a first indexingmethod that can handle moving averageand time warping.An index I is constructed asin [l,2] first.For each query,a transformation inT is given.不要放入翻译词典 再拿出来交给我好不好。囧