WO2014091712A1 - 可視化装置、可視化方法および可視化プログラム - Google Patents
可視化装置、可視化方法および可視化プログラム Download PDFInfo
- Publication number
- WO2014091712A1 WO2014091712A1 PCT/JP2013/007078 JP2013007078W WO2014091712A1 WO 2014091712 A1 WO2014091712 A1 WO 2014091712A1 JP 2013007078 W JP2013007078 W JP 2013007078W WO 2014091712 A1 WO2014091712 A1 WO 2014091712A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- attribute group
- evaluation index
- visualization
- data
- attribute
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
Definitions
- the present invention relates to a visualization device and a visualization method for making it possible to grasp an outline of a data structure, in particular, a data distribution, in order to examine preprocessing and analysis methods to be applied when performing data analysis on high-dimensional data.
- the present invention relates to a method and a visualization program.
- Data visualization technology is an indispensable technology for efficient data analysis work. With the development of sensing technology and information management technology, the need for data has become ever larger and more complex.
- the following scenes can be considered as typical application scenes of data visualization technology in the field of data analysis.
- PCP Parallel Coordinate Plot
- Principal component analysis and multidimensional scaling are useful in terms of preserving and visualizing information indicating the scattering of data points in a multidimensional space as much as possible.
- SPM and PCP are useful in that the relationship between specific minority dimensions existing in a high dimension is visualized as a whole with one figure.
- Patent Document 1 describes a method of classifying learning data having continuous numerical attributes by a decision tree based on data distribution characteristics to facilitate determination and change of a model structure.
- each node in the generated decision tree is divided into an objective function that is a data group related to a certain attribute and the remaining plural attributes. Is displayed as a scatter diagram with an explanatory function which is a data group related to.
- the attribute displayed as a scatter diagram is fixed to an objective variable and one explanatory variable. Therefore, the attribute to be displayed cannot be set for each node. Therefore, for example, when analyzing a data structure existing in high-dimensional data, the data structure that the user (analyzer) can overview is limited to a part of the data structure.
- the present invention provides a visualization apparatus, a visualization method, and a visualization method that can provide an overview of a characteristic data structure existing in high-dimensional data to be analyzed when a user analyzes high-dimensional data.
- the purpose is to provide a visualization program.
- the visualization apparatus relates to data related to a second attribute group including one or more attributes that is conditioned by the first attribute group including one or more attributes among the high-dimensional data to be visualized.
- An evaluation index calculation unit that calculates a value of an evaluation index that represents the degree of feature for each combination of the first attribute group and the second attribute group, and the value of the evaluation index based on a predetermined criterion
- a visualization processing unit that generates image information for presenting a combination of the first attribute group and the second attribute group, which is determined to be large.
- the visualization apparatus narrows down data having a characteristic structure in the high-dimensional data before performing the visualization process itself.
- the data is given in a matrix format of [sample ⁇ attribute].
- attributes and dimensions are synonymous.
- the visualization apparatus applies the visualization process itself to the narrowed data.
- As a method for narrowing down data having a characteristic structure consider focusing on data related to some other attribute (second attribute group) conditioned by some attribute (first attribute group).
- second attribute group conditioned by some attribute
- first attribute group it is assumed that there are a plurality of combinations of the first attribute group and the second attribute group. Therefore, it is desirable to extract as many combinations as possible from the viewpoint of visualization that allows an analyst to view as many data structures as possible.
- evaluation index for example, an amount such as a correlation coefficient or a degree of separation / mutual information when class information is given can be considered.
- Class information is an attribute representing the amount to be predicted in data analysis. For example, in marketing in CRM (Customer Relationship Management), an attribute representing a customer's purchase behavior corresponds to class information.
- the degree of separation is the accuracy rate of classification.
- the visualization apparatus provides an interface that allows an analyst to view the entire combination at a time as much as possible.
- a method for allowing the entire combination of attribute groups to be overviewed at a time there is a method of arranging graphs in a tree view (Tree View) format to be described later.
- the visualization method relates to data related to a second attribute group including one or more attributes that are conditioned by the first attribute group including one or more attributes among the high-dimensional data to be visualized. , Calculating the value of the evaluation index representing the degree of feature for each combination of the first attribute group and the second attribute group, and determined that the value of the evaluation index is large based on a predetermined criterion, Image information for presenting a combination of the first attribute group and the second attribute group is generated.
- the visualization program provides a computer with a second attribute group including one or more attributes conditioned by a first attribute group including one or more attributes among the high-dimensional data to be visualized.
- the process of calculating the value of the evaluation index representing the feature degree for each combination of the first attribute group and the second attribute group, and the value of the evaluation index based on a predetermined criterion A process of generating image information for presenting a combination of the first attribute group and the second attribute group, which is determined to be large, is executed.
- Embodiment 1 FIG. A first embodiment of the present invention will be described below with reference to the drawings.
- FIG. 1 is an explanatory diagram showing an example of the result of visualizing five-dimensional data with SPM.
- the sample represents, for example, each customer when the class information is purchase behavior.
- FIG. 1 shows a scatter diagram for any two-dimensional combination of the five dimensions.
- a decision tree is used as a method for narrowing down data. Also, since the classification problem is considered here, the degree of separation between “ ⁇ ” and “ ⁇ ” is considered as an index representing the feature level.
- FIG. 2 is an explanatory diagram showing an example of a decision tree created based on the five-dimensional data shown in FIG.
- the inequality in the decision tree shown in FIG. 2 represents a branch condition, and the leaf number represents “number of circles” or “number of x”.
- the visualization apparatus extracts “ ⁇ ” and “x” as nodes closer to the root of the tree, that is, nodes closer to the top of FIG. Therefore, for example, it can be read from the decision tree shown in FIG. 2 that the attribute V1 is the best separation between “ ⁇ ” and “x” as in the scatter diagram shown in FIG. Further, it can be read that the attribute V2 is the best separation between “ ⁇ ” and “x” for the data with the attribute V1 value of 0.02335 or more. Further, it can be read that the attribute V3 is the best separation between “ ⁇ ” and “x” for the data with the attribute V1 value less than 0.02335.
- the visualization apparatus provides an interface as shown in FIG. 3 as an output result so that the information regarding the classification of “ ⁇ ” and “x” can be overviewed at a time.
- FIG. 3 is an explanatory diagram showing an example of the output result of the visualization device for the five-dimensional data shown in FIG.
- three scatter diagrams are arranged so as to correspond to the structure of the decision tree shown in FIG.
- a dotted line in each scatter diagram shown in FIG. 3 is drawn at a position corresponding to the value of the left side of the inequality in each node shown in FIG.
- the scatter diagram B represents data related to the attribute V3, the attribute V5, and the attribute V4, which is limited to a sample having a small value of the attribute V1, that is, conditioned by the attribute V1.
- the attribute V1 corresponds to the first attribute group
- the attribute V3, the attribute V5, and the attribute V4 correspond to the second attribute group.
- the attribute V1 corresponds to the first attribute group
- the attribute V2, the attribute V4, and the attribute V5 correspond to the second attribute group.
- FIG. 3 includes three scatter plots, but the output result may include any number of scatter plots.
- the number of scatter plots may be changed according to the depth of the tree, or scatter corresponding to nodes that separate " ⁇ " and "x" well, that is, nodes with particularly large evaluation index values. Only the figure may be included in the output result. Note that whether or not the node sufficiently separates “O” and “X” may be determined based on, for example, whether or not the value of the evaluation index is greater than a predetermined reference.
- FIG. 4 is a block diagram showing the configuration of the first embodiment of the visualization apparatus according to the present invention.
- the visualization apparatus includes a data / parameter input unit 101, a decision tree creation unit 102, a decision tree storage unit 103, a tree view format scatter diagram creation unit 104, and an image output unit 105. .
- the data / parameter input unit 101 inputs high-dimensional data to be visualized, decision tree parameters, and scatter diagram parameters from the outside of the apparatus.
- the decision tree parameter is a parameter necessary when creating a decision tree.
- the scatter diagram parameter is a parameter necessary for image output.
- the data parameter input unit 101 inputs, for example, an index for evaluating the goodness of division, an algorithm for creating a decision tree, the depth of the tree, the minimum size of data belonging to a leaf, and the like as decision tree parameters.
- the data / parameter input unit 101 outputs the input data and parameters to the decision tree creation unit 102 and the tree view format scatter diagram creation unit 104.
- the decision tree creation unit 102 inputs data to be visualized and decision tree parameters from the data / parameter input unit 101.
- the decision tree creation unit 102 creates a decision tree as shown in FIG. 2 for the input data, according to the information of the input decision tree parameters. Note that numerical values such as “0.02335” and “ ⁇ 1.227” in the inequality shown in FIG. 2 are numerical values calculated by the decision tree creation unit 102 and are examples thereof.
- the decision tree creation unit 102 outputs information about the created decision tree to the decision tree storage unit 103.
- the decision tree storage unit 103 stores information regarding the decision tree input from the decision tree creation unit 102.
- the tree view scatter diagram creation unit 104 inputs data to be visualized and scatter diagram parameters from the data / parameter input unit 101.
- the tree view format scatter diagram creation unit 104 acquires information about the decision tree from the decision tree storage unit 103.
- the tree view format scatter diagram creation unit 104 scatters data to be visualized as shown in FIG. 1 in a tree view format as shown in FIG. 3 based on the scatter diagram parameters and information about the decision tree. Create a diagram.
- the tree view scatter diagram creation unit 104 outputs image information including the created scatter diagram to the image output unit 105.
- the scatter diagram parameters are parameters for designating, for example, the color and shape of the sample in the scatter diagram shown in FIG. 3, the presence / absence of axis labels and scales, and the presence / absence of dotted lines.
- the image output unit 105 is, for example, a display device or a printer.
- the image output unit 105 outputs image information including the tree view format scatter diagram input from the tree view format scatter diagram creation unit 104.
- the data / parameter input unit 101, the decision tree creation unit 102, and the tree view format scatter diagram creation unit 104 are realized by, for example, a computer that operates according to a visualization program.
- the CPU may read the visualization program and operate as the data / parameter input unit 101, the decision tree creation unit 102, and the tree view format scatter diagram creation unit 104 according to the program.
- the data / parameter input unit 101, the decision tree creation unit 102, and the tree view format scatter diagram creation unit 104 may be realized by separate hardware.
- the decision tree storage unit 103 is realized by a storage device such as a memory provided in the visualization device.
- FIG. 5 is a flowchart showing the operation of the first embodiment of the visualization apparatus according to the present invention.
- the data / parameter input unit 101 inputs high-dimensional data to be visualized, decision tree parameters, and scatter diagram parameters (step S101). For example, the analyst inputs the data and the parameters to the data / parameter input unit 101.
- the data / parameter input unit 101 outputs the high-dimensional data to be visualized and the decision tree parameters to the decision tree creation unit 102.
- the data / parameter input unit 101 outputs high-dimensional data to be visualized and scatter diagram parameters to the tree view format scatter diagram creation unit 104.
- the decision tree creation unit 102 creates a decision tree according to the data to be visualized input from the data / parameter input unit 101 and the information of the decision tree parameter, and stores information about the created decision tree in the decision tree storage unit 103. Store (step S102).
- the tree view format scatter diagram creation unit 104 is configured as shown in FIG. A scatter diagram as shown, that is, a scatter diagram arranged in a tree view format is created (step S103).
- the tree view scatter diagram creation unit 104 outputs image information including the created scatter diagram to the image output unit 105.
- the image output unit 105 outputs image information including a tree view format scatter diagram (step S104).
- data having a characteristic structure is narrowed down in high-dimensional data before performing the visualization process itself.
- a tree view format created based on information about the decision tree used to narrow down the data so that more characteristic data structures existing in high-dimensional data can be viewed more at once.
- Output a scatter plot of. Accordingly, when analyzing high-dimensional data, the analyst can grasp a plurality of characteristic data structures existing in the data. As a result, the analyst can make a guide as to what pre-processing and analysis method should be applied.
- samples having a characteristic data structure are extracted in advance and analyzed separately as preprocessing when analyzing high-dimensional data. It becomes possible. In addition, it is possible to provide information for determining the number of mixtures and a model in each component when a multimodal model such as a mixed normal distribution is applied to data as analysis means.
- the visualization device includes the image output unit.
- the image output unit may not be included in the visualization device.
- the tree view scatter diagram creation unit 104 may output image information to an external display or printer that can communicate with the visualization device. According to such a configuration, the configuration of the visualization device can be further simplified.
- the degree of separation is used as the evaluation index representing the feature degree.
- a correlation coefficient may be used.
- entropy may be used as an evaluation index representing the feature level.
- the case where the scatter diagram is displayed in the tree view format is described.
- the PCP may be displayed in the tree view format.
- the case where the decision tree creation unit 102 creates a decision tree using a binary tree has been described.
- the decision tree creation unit 102 creates a decision tree using a multi-tree other than the binary tree. You may make it do.
- the tree view format scatter diagram creation unit 104 may create a tree view format scatter diagram corresponding to the configuration of the decision tree.
- the present invention is applicable to all scenes where knowledge is discovered from high-dimensional data.
- the visualization device according to the present invention is used. Can be used. In that case, it is possible to overview in advance whether or not there is a characteristic structure between the data relating to the type of the specific cause of failure and the data relating to the type of sensor at the examination stage of the analysis method.
- the visualization apparatus according to the present invention can be used in the same way in the case of investigating the relationship between customer personal information and purchasing behavior in CRM marketing.
- FIG. 6 is a block diagram showing the minimum configuration of the visualization apparatus according to the present invention.
- the visualization device includes a second attribute including one or more attributes that is conditioned by a first attribute group including one or more attributes among the high-dimensional data to be visualized.
- the evaluation index calculation unit 11 calculates the value of the evaluation index representing the degree of feature for each combination of the first attribute group and the second attribute group.
- Visualization processing for generating image information for presenting a combination of the first attribute group and the second attribute group, which has been determined that the value of the evaluation index is large based on a predetermined criterion And a unit 12 (corresponding to the tree view scatter diagram creation unit 104 shown in FIG. 4).
- a visualization device in which the evaluation index calculation unit 11 uses a binary tree as a criterion for conditioning the second attribute group with the first attribute group.
- a visualization apparatus in which the evaluation index calculation unit 11 uses the correct rate of class classification as an evaluation index representing the degree of feature when class information is given regarding data to be visualized.
- a visualization apparatus that uses entropy as an evaluation index representing the degree of feature when the evaluation index calculation unit 11 is given class information regarding data to be visualized.
- the evaluation index representing the characteristic degree of the data structure can be expressed as a numerical value, the comparison of the evaluation index for each combination of the first attribute group and the second attribute group is more performed. Can be done accurately.
- Image information in which a two-dimensional scatter diagram corresponding to the combination of the first attribute group and the second attribute group, which is determined by the visualization processing unit 12 to have a large evaluation index value, is arranged in a tree view format.
- Visualization device to output.
- the visualization processing unit 12 outputs image information in which parallel coordinate plots corresponding to combinations of the first attribute group and the second attribute group, which are determined to have a large evaluation index value, are arranged in a tree view format. Visualization device.
- the analyst can overview more characteristic data structures existing in the high-dimensional data to be visualized at a time.
- combinations of attribute groups having larger evaluation index values are displayed at positions closer to the top of the tree view, the analyst can easily interpret the visualized contents.
- only scatter diagrams corresponding to combinations of attribute groups with particularly large evaluation index values can be presented to the user, even if the number of data and the number of dimensions of the high-dimensional data to be visualized increase, There is no complication.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
以下、本発明の第1の実施形態を図面を参照して説明する。
12 可視化処理部
101 データ・パラメータ入力部
102 決定木作成部
103 決定木記憶部
104 ツリービュー形式散布図作成部
105 画像出力部
Claims (10)
- 可視化対象となる高次元データのうち、1つまたは複数の属性を含む第一の属性群で条件付けした、1つまたは複数の属性を含む第二の属性群に関するデータについて、前記第一の属性群と前記第二の属性群との組み合わせそれぞれに対して特徴度合いを表す評価指標の値を算出する評価指標算出部と、
予め定められた基準をもとに評価指標の値が大きいと判断した、第一の属性群と第二の属性群との組み合わせを提示するための画像情報を生成する可視化処理部とを備えた
ことを特徴とする可視化装置。 - 評価指標算出部は、第一の属性群で第二の属性群を条件付けする基準として二分木を用いる
請求項1に記載の可視化装置。 - 評価指標算出部は、第一の属性群で第二の属性群を条件付けする基準として多分木を用いる
請求項1に記載の可視化装置。 - 評価指標算出部は、特徴度合いを表す評価指標として相関係数を用いる
請求項1から請求項3のうちのいずれか1項に記載の可視化装置。 - 評価指標算出部は、可視化対象となるデータに関してクラス情報が与えられているとき、特徴度合いを表す評価指標として、クラス分類の正解率を用いる
請求項1から請求項3のうちのいずれか1項に記載の可視化装置。 - 評価指標算出部は、可視化対象となるデータに関してクラス情報が与えられているとき、特徴度合いを表す評価指標として、エントロピーを用いる
請求項1から請求項3のうちのいずれか1項に記載の可視化装置。 - 可視化処理部は、評価指標の値が大きいと判断した、第一の属性群と第二の属性群との組み合わせに対応する二次元散布図をツリービュー形式で並べた画像情報を出力する
請求項1から請求項6のうちのいずれか1項に記載の可視化装置。 - 可視化処理部は、評価指標の値が大きいと判断した、第一の属性群と第二の属性群との組み合わせに対応する平行座標プロットをツリービュー形式で並べた画像情報を出力する
請求項1から請求項6のうちのいずれか1項に記載の可視化装置。 - 可視化対象となる高次元データのうち、1つまたは複数の属性を含む第一の属性群で条件付けした、1つまたは複数の属性を含む第二の属性群に関するデータについて、前記第一の属性群と前記第二の属性群との組み合わせそれぞれに対して特徴度合いを表す評価指標の値を算出し、
予め定められた基準をもとに評価指標の値が大きいと判断した、第一の属性群と第二の属性群との組み合わせを提示するための画像情報を生成する
ことを特徴とする可視化方法。 - コンピュータに、
可視化対象となる高次元データのうち、1つまたは複数の属性を含む第一の属性群で条件付けした、1つまたは複数の属性を含む第二の属性群に関するデータについて、前記第一の属性群と前記第二の属性群との組み合わせそれぞれに対して特徴度合いを表す評価指標の値を算出する処理と、
予め定められた基準をもとに評価指標の値が大きいと判断した、第一の属性群と第二の属性群との組み合わせを提示するための画像情報を生成する処理とを実行させる
ための可視化プログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/652,153 US10013469B2 (en) | 2012-12-13 | 2013-12-03 | Visualization device, visualization method and visualization program |
JP2014551875A JP6330665B2 (ja) | 2012-12-13 | 2013-12-03 | 可視化装置、可視化方法および可視化プログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012272413 | 2012-12-13 | ||
JP2012-272413 | 2012-12-13 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2014091712A1 true WO2014091712A1 (ja) | 2014-06-19 |
Family
ID=50934020
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/007078 WO2014091712A1 (ja) | 2012-12-13 | 2013-12-03 | 可視化装置、可視化方法および可視化プログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US10013469B2 (ja) |
JP (1) | JP6330665B2 (ja) |
WO (1) | WO2014091712A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016079909A1 (ja) * | 2014-11-19 | 2016-05-26 | 日本電気株式会社 | 可視化装置、可視化方法および可視化プログラム |
JP2019174960A (ja) * | 2018-03-27 | 2019-10-10 | 西日本電信電話株式会社 | データ前処理装置、データ前処理方法及びデータ前処理プログラム |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11244401B2 (en) * | 2015-10-30 | 2022-02-08 | Hartford Fire Insurance Company | Outlier system for grouping of characteristics |
US10628456B2 (en) | 2015-10-30 | 2020-04-21 | Hartford Fire Insurance Company | Universal analytical data mart and data structure for same |
US10942929B2 (en) | 2015-10-30 | 2021-03-09 | Hartford Fire Insurance Company | Universal repository for holding repeatedly accessible information |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008299363A (ja) * | 2007-05-29 | 2008-12-11 | Ochanomizu Univ | データベースを可視化するための画像生成方法及びプログラム並びにデータベース表示装置 |
Family Cites Families (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6750864B1 (en) * | 1999-11-15 | 2004-06-15 | Polyvista, Inc. | Programs and methods for the display, analysis and manipulation of multi-dimensional data implemented on a computer |
ATE488806T1 (de) | 2000-03-16 | 2010-12-15 | Poly Vista Inc | System und methode zur analyse einer anfrage und zur erzeugung von antworten und damit verbundenen fragen |
US6661362B2 (en) * | 2000-09-11 | 2003-12-09 | Broadcom Corporation | Methods and systems for high speed quantizers |
US6819344B2 (en) * | 2001-03-12 | 2004-11-16 | Microsoft Corporation | Visualization of multi-dimensional data having an unbounded dimension |
JP3897169B2 (ja) | 2002-11-07 | 2007-03-22 | 富士電機ホールディングス株式会社 | 決定木生成方法およびモデル構造生成装置 |
US7557805B2 (en) * | 2003-04-01 | 2009-07-07 | Battelle Memorial Institute | Dynamic visualization of data streams |
US7797320B2 (en) * | 2005-03-15 | 2010-09-14 | Hyperion Solutions Corporation | Dimensionality reduction |
US20070005582A1 (en) * | 2005-06-17 | 2007-01-04 | Honeywell International Inc. | Building of database queries from graphical operations |
US20080071843A1 (en) * | 2006-09-14 | 2008-03-20 | Spyridon Papadimitriou | Systems and methods for indexing and visualization of high-dimensional data via dimension reorderings |
US20080111826A1 (en) * | 2006-11-14 | 2008-05-15 | Microsoft Corporation | Displaying resources using multiple visualization dimensions |
US8640056B2 (en) * | 2007-07-05 | 2014-01-28 | Oracle International Corporation | Data visualization techniques |
US8935249B2 (en) * | 2007-06-26 | 2015-01-13 | Oracle Otc Subsidiary Llc | Visualization of concepts within a collection of information |
US8446412B2 (en) * | 2008-06-26 | 2013-05-21 | Microsoft Corporation | Static visualization of multiple-dimension data trends |
WO2010064939A1 (en) * | 2008-12-05 | 2010-06-10 | Business Intelligence Solutions Safe B.V. | Methods, apparatus and systems for data visualization and related applications |
US20100194778A1 (en) * | 2009-01-30 | 2010-08-05 | Microsoft Corporation | Projecting data dimensions on a visualization data set |
US20110029926A1 (en) * | 2009-07-30 | 2011-02-03 | Hao Ming C | Generating a visualization of reviews according to distance associations between attributes and opinion words in the reviews |
US7812838B1 (en) * | 2010-01-12 | 2010-10-12 | Rana Ian | Typed data graph visualization system in three dimensions |
CA2709507A1 (en) * | 2010-07-14 | 2012-01-14 | Oculus Info Inc. | System and method for visualizing multi-dimensional data using shape attributes |
US20120013619A1 (en) * | 2010-07-14 | 2012-01-19 | Richard Brath | System and method for visualizing multi-dimensional data using shape attributes |
US9799004B2 (en) * | 2010-07-30 | 2017-10-24 | Avaya Inc. | System and method for multi-model, context-aware visualization, notification, aggregation and formation |
US8615511B2 (en) * | 2011-01-22 | 2013-12-24 | Operational Transparency LLC | Data visualization interface |
US9342579B2 (en) * | 2011-05-31 | 2016-05-17 | International Business Machines Corporation | Visual analysis of multidimensional clusters |
US9064245B2 (en) * | 2012-02-22 | 2015-06-23 | Hewlett-Packard Development Company, L.P. | Generating a calendar graphical visualization including pixels representing data records containing user feedback |
-
2013
- 2013-12-03 JP JP2014551875A patent/JP6330665B2/ja active Active
- 2013-12-03 WO PCT/JP2013/007078 patent/WO2014091712A1/ja active Application Filing
- 2013-12-03 US US14/652,153 patent/US10013469B2/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008299363A (ja) * | 2007-05-29 | 2008-12-11 | Ochanomizu Univ | データベースを可視化するための画像生成方法及びプログラム並びにデータベース表示装置 |
Non-Patent Citations (3)
Title |
---|
HIROSHI ISHIKAWA: "Data Mining to Shugochi -Kiso kara Web, Social Media made", MITSUAKI NANJO, 15 July 2012 (2012-07-15), pages 199 - 211 * |
UNJU TEI ET AL.: "Teijigen Plot no Shugo ni yoru Kojigen Data Kashika no Ichi Shuho (1)", 2012 NENDO ANNUAL CONFERENCE OF JSAI (DAI 26 KAI) RONBUNSHU [ CD-ROM ] 2012 NENDO ANNUAL CONFERENCE OF JSAI (DAI 26 KAI) RONBUNSHU, 12 June 2012 (2012-06-12), pages 1 - 4 * |
YUICHI IIZUKA ET AL.: "A Study on Data Visualization Based on Category and Characteristics", IPSJ SIG NOTES, vol. 99, no. 6, 23 January 1999 (1999-01-23), pages 7 - 12 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016079909A1 (ja) * | 2014-11-19 | 2016-05-26 | 日本電気株式会社 | 可視化装置、可視化方法および可視化プログラム |
JP2019174960A (ja) * | 2018-03-27 | 2019-10-10 | 西日本電信電話株式会社 | データ前処理装置、データ前処理方法及びデータ前処理プログラム |
Also Published As
Publication number | Publication date |
---|---|
US10013469B2 (en) | 2018-07-03 |
JP6330665B2 (ja) | 2018-05-30 |
US20150339364A1 (en) | 2015-11-26 |
JPWO2014091712A1 (ja) | 2017-01-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6330665B2 (ja) | 可視化装置、可視化方法および可視化プログラム | |
Kaya et al. | Process capability analyses based on fuzzy measurements and fuzzy control charts | |
Sawatsky et al. | Partial least squares regression in the social sciences | |
US9299173B2 (en) | Automatic selection of different visualizations for the organization of multivariate data | |
Pagliosa et al. | Projection inspector: Assessment and synthesis of multidimensional projections | |
Heimerl et al. | embcomp: Visual interactive comparison of vector embeddings | |
Halim et al. | Quantifying and optimizing visualization: An evolutionary computing-based approach | |
Giannetti et al. | A novel variable selection approach based on co-linearity index to discover optimal process settings by analysing mixed data | |
Young et al. | Descriptive statistics, graphs, and visualisation | |
JP5882272B2 (ja) | 資料評価プログラムおよび資料評価装置 | |
Areosa et al. | Explaining the performance of black box regression models | |
KR101774712B1 (ko) | 특허청구범위 분석 방법 및 그 장치 | |
Bandaru et al. | Trend mining: A visualization technique to discover variable trends in the objective space | |
Steed et al. | QVis: A Visual Analytics Tool for Exploring Noise and Errors in Quantum Computing Systems | |
Dankov et al. | Extended conceptual framework for business analytics supporting innovations | |
Chandrasekar et al. | Data Exploratory Analysis for Classification in Machine Learning Algorithms | |
Pimentel et al. | Subgroup mining for performance analysis of regression models | |
JP7455086B2 (ja) | 機械学習モデル特性可視化装置、機械学習モデル特性可視化方法、及び機械学習モデル特性可視化プログラム | |
WO2012046436A1 (ja) | 文書分析装置、文書分析方法および文書分析プログラム | |
Schader et al. | LayoutExOmizer: Interactive exploration and optimization of 2d data layouts | |
Ruiz-Gazen et al. | Detecting outliers in compositional data using Invariant Coordinate Selection | |
Cummings et al. | Detecting Dataset Drift and Non-IID Sampling via k-Nearest Neighbors | |
Nanaware et al. | Exploratory data analysis using dimension reduction | |
Diago et al. | Analyzing facial expressions with fuzzy quantification theory II: indefinite generalized eigenvalue problem | |
CN117540107B (zh) | 一种数据处理方法、装置、电子设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13861886 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014551875 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14652153 Country of ref document: US |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13861886 Country of ref document: EP Kind code of ref document: A1 |