WO2016206188A1 - 输出对象的数据处理方法、装置、设备及非易失性计算机存储介质 - Google Patents
输出对象的数据处理方法、装置、设备及非易失性计算机存储介质 Download PDFInfo
- Publication number
- WO2016206188A1 WO2016206188A1 PCT/CN2015/088391 CN2015088391W WO2016206188A1 WO 2016206188 A1 WO2016206188 A1 WO 2016206188A1 CN 2015088391 W CN2015088391 W CN 2015088391W WO 2016206188 A1 WO2016206188 A1 WO 2016206188A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- output
- objects
- output objects
- similarity
- obtaining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9038—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Definitions
- the present invention relates to data processing technologies, and in particular, to a data processing method, apparatus, device, and non-volatile computer storage medium for output objects.
- the ratio of the intersection of the feature attributes of the two output objects and the union of the feature attributes of the two output objects can only represent a one-way relationship between the two output objects, the obtained two-two output objects are similar. Degree is not the true similarity of the two output objects, resulting in a reduction in the reliability of the output object visualization.
- aspects of the present invention provide a data processing method, apparatus, and apparatus for outputting an object, and a non-volatile computer storage medium for improving reliability of output object visualization.
- An aspect of the present invention provides a data processing method for an output object, including:
- Each of the output objects is visually output according to the similarity matrix.
- the feature attributes of the output object include:
- the number of people corresponding to the search keyword to which the output object belongs is the number of people corresponding to the search keyword to which the output object belongs.
- the similarity matrix is obtained according to the similarity of the two output objects.
- Each of the output objects is output according to an output position of each of the output objects.
- Each of the at least one random position of each of the output objects is an initial position of the output object, and an iterative process is performed according to the theoretical distance of the two output objects to obtain the output object. Iterative position
- Each of the output objects is output according to an output position of each of the output objects.
- any possible implementation manner further provide an implementation manner, where calculating the distance between the two output objects according to the similarity matrix, including:
- the N-th power of the ratio of the similarity between 1 and the two output objects is taken as the distance between the two output objects; if the similarity between the two output objects is 0, The Nth power of the ratio of the similarity between 1 and the smallest two-output object is taken as the distance between the two output objects; N is a number greater than 0 and less than 1.
- N 0.5
- each of the output objects is visually outputted in a two-dimensional plane space or a three-dimensional space.
- Another aspect of the present invention provides a data processing apparatus for outputting an object, including:
- An obtaining unit configured to acquire feature attributes of each of the at least two output objects
- An analyzing unit configured to obtain, according to the feature attribute of each output object, an intersection of feature attributes of the two output objects
- the analyzing unit is further configured to obtain a similarity matrix according to an intersection of a feature attribute of each output object and a feature attribute of the two output objects;
- an output unit configured to visually output each of the output objects according to the similarity matrix.
- the characteristic attributes of the output object include:
- the number of people corresponding to the search keyword to which the output object belongs is the number of people corresponding to the search keyword to which the output object belongs.
- the similarity matrix is obtained according to the similarity of the two output objects.
- Each of the output objects is output according to an output position of each of the output objects.
- Each of the at least one random position of each of the output objects is an initial position of the output object, and an iterative process is performed according to the theoretical distance of the two output objects to obtain the output object. Iterative position
- Each of the output objects is output according to an output position of each of the output objects.
- the N-th power of the ratio of the similarity between 1 and the two output objects is taken as the distance between the two output objects; if the similarity between the two output objects is 0, The Nth power of the ratio of the similarity between 1 and the smallest two-output object is taken as the distance between the two output objects; N is a number greater than 0 and less than 1.
- N 0.5
- each of the output objects is visually outputted in a two-dimensional plane space or a three-dimensional space.
- an apparatus comprising:
- One or more processors are One or more processors;
- One or more programs the one or more programs being stored in the memory, when executed by the one or more processors:
- Each of the output objects is visually output according to the similarity matrix.
- a nonvolatile computer storage medium storing one or more programs when the one or more programs are executed by a device causes The device:
- Each of the output objects is visually output according to the similarity matrix.
- the embodiment of the present invention obtains the feature attribute of each output object of at least two output objects, and further obtains the intersection of the feature attributes of the two output objects according to the feature attribute of each output object, and According to the characteristics of each output object Obtaining an intersection of the attribute and the feature attribute of the pair of output objects, obtaining a similarity matrix, so that each of the output objects can be visually output according to the similarity matrix, because according to the characteristic attribute of each output object and two
- the similarity matrix obtained by the intersection of the feature attributes of the two output objects can represent the bidirectional relationship between the two output objects, so that the similarity of the obtained two output objects is the true similarity of the two output objects, thereby improving The reliability of the output object visualization.
- each of the at least one random position of each output object is the initial position of the output object, and the theoretical distance of the two output objects is performed multiple times.
- the iterative processing makes it possible to obtain an optimal result, which can further improve the reliability of the output object visualization.
- the specific seed is preset, so that the result of each iterative process in the multiple iterative processes performed is relatively stable, and the stability of the output object visualization can be effectively improved.
- the output position interval of each output object is reduced, and the reliability of the output object visualization can be further improved.
- the technical solution provided by the present invention can effectively improve the user experience.
- FIG. 1 is a schematic flowchart of a data processing method of an output object according to an embodiment of the present invention
- FIG. 2 is a schematic structural diagram of a data processing apparatus for outputting an object according to another embodiment of the present invention.
- the terminals involved in the embodiments of the present invention may include, but are not limited to, a mobile phone, a personal digital assistant (PDA), a wireless handheld device, a tablet computer, and a personal computer (Personal Computer, PC). ), MP3 player, MP4 player, wearable device (for example, smart glasses, smart watches, smart bracelets, etc.).
- PDA personal digital assistant
- PC Personal Computer
- FIG. 1 is a schematic flowchart of a data processing method of an output object according to an embodiment of the present invention, as shown in FIG. 1 .
- the so-called output object refers to the object that can be visualized, for example, to promote the relationship between the product and other promotional products, or to promote the relationship between the product and media attributes, population attributes and geographical attributes, and so on.
- the so-called promotion product refers to a network promotion method for promoting user information, which may include, but is not limited to, search promotion, network promotion, brand advertisement, Baidu alliance, advertising butler, Baidu statistics, Baidu commercial bridge, etc., this embodiment This is not particularly limited.
- intersection of the feature attributes of the two output objects refers to the feature attributes common to the two output objects.
- each matrix element included in the similarity matrix may represent a similarity of a pair of output objects, and a matrix element on the main diagonal line may represent an output object and itself. Similarity.
- the similarity of each output object to itself can be specified in advance, for example, 0 or 1.
- the similarity matrix T can be in the form of:
- T ij (i ⁇ j) represents the similarity of a pair of output objects
- execution entities of 101 to 104 may be applications located in the local terminal, or may be plug-ins or software development kits (SDKs) installed in applications located in the local terminal.
- the functional unit may also be a processing engine located in the network side server, or may be a distributed system located on the network side, which is not specifically limited in this embodiment.
- the application may be a local application (nativeApp) installed on the terminal, or may be a web application (webApp) of the browser on the terminal, which is not limited in this embodiment.
- the feature attribute of the output object may include, but is not limited to, the following attributes:
- the number of people corresponding to the search keyword to which the output object belongs is the number of people corresponding to the search keyword to which the output object belongs.
- the ratio of the square of the intersection of the feature attributes of the two output objects to the product of the feature attributes of the two output objects is obtained.
- the similarity between the two objects is outputted, and further, the similarity matrix can be obtained according to the similarity of the two output objects.
- intersection of the feature attributes of the two output objects and the ratio of each of the two output objects are multiplied to obtain the similarity of the two output objects. degree.
- the similarity matrix is formed by the similarity of the two output objects.
- the similarity obtained according to the intersection of the feature attribute of each output object and the feature attribute of the two output objects can represent the bidirectional relationship between the two output objects, and thus the similarity of the obtained two output objects It is the true similarity between the two output objects, which can more accurately represent the relationship between the two output objects.
- the similarity matrix may be subjected to Singular Value Decomposition (SVD) processing to obtain an decomposition matrix, and then, According to the decomposition matrix, an output position of each of the output objects is obtained. Then, each of the output objects is output according to the output position of each of the output objects.
- Singular Value Decomposition Singular Value Decomposition
- the decomposition in addition to obtaining the output position of each output object, that is, the coordinate value in the output space defined by the specified coordinate axis, according to the decomposition matrix, the decomposition may be further performed according to the decomposition. a matrix that obtains the specified sitting in the output space The contribution of the target axis for output to the user.
- the theoretical distance between the two output objects may be calculated according to the similarity matrix, and further, according to the preset at least one specific seed. At least one random location of each of the output objects is obtained using a random function. Then, each of the at least one random position of each of the output objects is an initial position of the output object, and an iterative process is performed according to the theoretical distance of the two output objects to obtain each output. The iterative position of the object, and then the output position of each of the output objects is obtained according to the iteration position of each of the output objects. Then, each of the output objects is output according to the output position of each of the output objects.
- the so-called random location of the output object is not a true random location, but is actually a pseudo-random location.
- the pseudo-random function is based on a seed random function as the initial condition, and then generated by a certain algorithm without iteration. In this way, the specific seed is set in advance, so that the result of each iteration processing in the multiple iteration processing performed is relatively stable, and the stability of the output object visualization can be effectively improved.
- the ratio of the similarity between the 1 and the two output objects may be specifically used as the distance between the two output objects. This traditional way of defining distances, the accuracy of the obtained distance is very poor, especially when the matrix elements are included in a similarity matrix. When the difference between the values is large, the output position of each output object to be output is very large, which results in a decrease in the reliability of the visualization of the output object.
- the N-th power of the ratio of the similarity between 1 and the two output objects is taken as the distance between the two output objects; If the similarity between the two output objects is 0, the N-th power of the ratio of the similarity between 1 and the smallest two-output object is taken as the distance between the two output objects; N is a number greater than 0 and less than 1.
- N can take a value of 0.5.
- each of the output objects may be in an output space such as a two-dimensional plane space or a three-dimensional space according to the similarity matrix.
- Visualize the output thus, by visually outputting each of the output objects on the output space, it is possible to clearly determine the relationship between each of the output objects.
- the feature attributes of each of the at least two output objects are obtained, and then the intersection of the feature attributes of the two output objects is obtained according to the feature attributes of each of the output objects, and according to each Obtaining an intersection of a feature attribute of the output object and a feature attribute of the pair of output objects, obtaining a similarity matrix, so that each output object can be visually output according to the similarity matrix, because each output object is
- the similarity matrix obtained by the intersection of the feature attribute and the feature attribute of the two output objects can represent the bidirectional relationship between the two output objects, so that the similarity of the obtained two output objects is the true similarity of the two output objects. Degree, which improves the reliability of the output object visualization.
- each of the at least one random position is an initial position of the output object, and the iterative process is performed according to the theoretical distance of the two output objects, so that an optimal result can be obtained, and the reliability of the output object visualization can be further improved.
- the specific seed is preset, so that the result of each iterative process in the multiple iterative processes performed is relatively stable, and the stability of the output object visualization can be effectively improved.
- the output position interval of each output object is reduced, and the reliability of the output object visualization can be further improved.
- the technical solution provided by the present invention can effectively improve the user experience.
- the data processing apparatus of the output object of the present embodiment may include an acquisition unit 21, an analysis unit 22, and an output unit 23.
- the obtaining unit 21 is configured to acquire feature attributes of each of the at least two output objects
- the analyzing unit 22 is configured to obtain an intersection of the feature attributes of the two output objects according to the feature attributes of each of the output objects.
- the analyzing unit 22 is further configured to obtain a similarity matrix according to an intersection of a feature attribute of each output object and a feature attribute of the two output objects, and an output unit 23 configured to use, according to the similarity matrix,
- Each of the output objects is visually output.
- part or all of the execution body of the data processing apparatus of the output object provided by this embodiment may be an application located in the local terminal, or may be a plug-in or software development tool disposed in an application located in the local terminal.
- the functional unit such as a software development kit (SDK)
- SDK software development kit
- the application may be a local application (nativeApp) installed on the terminal, or may be a web application (webApp) of the browser on the terminal, which is not limited in this embodiment.
- the feature attribute of the output object may include, but is not limited to, the following attributes:
- the number of people corresponding to the search keyword to which the output object belongs is the number of people corresponding to the search keyword to which the output object belongs.
- the analyzing unit 22 may be specifically configured to use, according to a product of an intersection of feature attributes of the two output objects and a product of the feature attributes of the two output objects. And comparing the similarities of the two output objects; and obtaining the similarity matrix according to the similarity of the two output objects.
- the output unit 23 may be specifically configured to perform SVD processing on the similarity matrix to obtain an decomposition matrix, and obtain the foregoing according to the decomposition matrix. An output position of each output object; and outputting each of the output objects according to an output position of each of the output objects.
- the output unit 23 may be configured to calculate a theoretical distance of the two output objects according to the similarity matrix; obtain, by using a random function, at least one random location of each output object according to the preset at least one specific seed; Each of the at least one random position of each output object is an initial position of the output object, and iteratively processing is performed according to the theoretical distance of the two output objects to obtain an iteration position of each output object; An iteration position of each of the output objects, an output position of each of the output objects is obtained; and each of the output objects is output according to an output position of each of the output objects.
- the output unit 23 may be specifically configured to use a ratio of similarities between 1 and two output objects as the distance between the two output objects.
- the output unit 23 may be specifically configured to: if the similarity between the two output objects is not 0, the N-th power of the ratio of the similarity between 1 and the two output objects is taken as The distance between the two output objects; if the similarity between the two output objects is 0, the N-th power of the ratio of the similarity between 1 and the minimum two output objects is taken as the distance between the two output objects; N is greater than 0 and a number less than 1.
- N can take a value of 0.5.
- the output unit 23 may be specifically configured to: perform the output on the two-dimensional plane space or the three-dimensional space according to the similarity matrix. The object is visually output.
- the feature attribute of each of the at least two output objects is obtained by the acquiring unit, and then obtained by the analyzing unit according to the feature attribute of each output object.
- Visualizing the output object the bidirectional relationship between the two output objects can be represented by the similarity matrix obtained according to the intersection of the feature attribute of each output object and the feature attributes of the two output objects
- the similarity of the obtained two-output objects is the true similarity between the two output objects, thereby improving the reliability of the output object visualization.
- each of the at least one random position of each output object is the initial position of the output object, and the theoretical distance of the two output objects is performed multiple times.
- the iterative processing makes it possible to obtain an optimal result, which can further improve the reliability of the output object visualization.
- the specific seed is preset, so that the result of each iterative process in the multiple iterative processes performed is relatively stable, and the stability of the output object visualization can be effectively improved.
- the output position interval of each output object is reduced, and the reliability of the output object visualization can be further improved.
- the technical solution provided by the present invention can effectively improve the user experience.
- the disclosed system, apparatus, and method may be implemented in other manners.
- the device embodiment described above For example, the division of the unit is only a logical function division, and the actual implementation may have another division manner, for example, multiple units or components may be combined or may be integrated into another system, or some Features can be ignored or not executed.
- the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
- the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
- each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
- the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
- the above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium.
- the above software functional unit is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform the methods of the various embodiments of the present invention. Part of the steps.
- the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Computational Linguistics (AREA)
Abstract
一种输出对象的数据处理方法、装置、设备及非易失性计算机存储介质。该方法通过获取至少两个输出对象中每个输出对象的特征属性(101),进而根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集(102),以及根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵(103),使得能够根据所述相似度矩阵,对所述每个输出对象进行可视化输出(104),由于根据每个输出对象的特征属性和两两输出对象的特征属性的交集所获得的相似度矩阵,能够表示两两输出对象之间的双向关系,使得所获得的两两输出对象的相似度是两两输出对象真正的相似度,从而提高了输出对象可视化的可靠性。
Description
本申请要求了申请日为2015年06月23日,申请号为201510349469.0发明名称为“输出对象的数据处理方法及装置”的中国专利申请的优先权。
本发明涉及数据处理技术,特别涉及一种输出对象的数据处理方法、装置、设备及非易失性计算机存储介质。
随着互联网的不断发展,不同行业的用户所面临的数据量越来越大,从这些海量数据中找到他们所关心的内容即输出对象,并且可视化是一个很大的挑战。例如,广告商可以通过选择不同的推广产品,来向公众提供推广服务,其需要知道推广产品与其他推广产品之间的关系,或者推广产品与媒体属性、人口属性和地域属性之间的关系,等等,并且还需要将这些多维度关系展现在一个如二维平面空间等较低维度空间上。在展现的过程中,可以将两两输出对象的特征属性的交集与该两两输出对象的特征属性的并集的比值,作为该两两输出对象的相似度。
然而,由于两两输出对象的特征属性的交集与该两两输出对象的特征属性的并集的比值只能够表示两两输出对象之间的单向关系,使得所获得的两两输出对象的相似度并不是两两输出对象真正的相似度,从而导致了输出对象可视化的可靠性的降低。
发明内容
本发明的多个方面提供一种输出对象的数据处理方法、装置、设备及非易失性计算机存储介质,用以提高输出对象可视化的可靠性。
本发明的一方面,提供一种输出对象的数据处理方法,包括:
获取至少两个输出对象中每个输出对象的特征属性;
根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;
根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;
根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述输出对象的特征属性包括:
输出对象所属搜索关键词所对应的人数。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵,包括:
根据两两输出对象的特征属性的交集的平方与该两两输出对象的特征属性的乘积的比值,获得该两两输出对象的相似度;
根据该两两输出对象的相似度,获得所述相似度矩阵。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述根据所述相似度矩阵,对所述每个输出对象进行可视化输出,包括:
对所述相似度矩阵进行SVD处理,以获得分解矩阵;
根据所述分解矩阵,获得所述每个输出对象的输出位置;
根据所述每个输出对象的输出位置,输出所述每个输出对象。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述根据所述相似度矩阵,对所述每个输出对象进行可视化输出,包括:
根据所述相似度矩阵,计算两两输出对象的理论距离;
根据预先设置的至少一个特定种子,利用随机函数,获得所述每个输出对象的至少一个随机位置;
分别以所述每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行迭代处理,以获得所述每个输出对象的迭代位置;
根据所述每个输出对象的迭代位置,获得所述每个输出对象的输出位置;
根据所述每个输出对象的输出位置,输出所述每个输出对象。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述根据所述相似度矩阵,计算两两输出对象的距离,包括:
将1与两两输出对象的相似度的比值,作为该两两输出对象的距离;或者
若两两输出对象的相似度不为0,将1与该两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;若两两输出对象的相似度为0,将1与最小的两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;N为大于0且小于1的数。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,N为0.5。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述根据所述相似度矩阵,对所述每个输出对象进行可视化输出,包括:
根据所述相似度矩阵,在二维平面空间或三维立体空间上,对所述每个输出对象进行可视化输出。
本发明的另一方面,提供一种输出对象的数据处理装置,包括:
获取单元,用于获取至少两个输出对象中每个输出对象的特征属性;
分析单元,用于根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;
所述分析单元,还用于根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;
输出单元,用于根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,
所述输出对象的特征属性包括:
输出对象所属搜索关键词所对应的人数。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述分析单元,具体用于
根据两两输出对象的特征属性的交集的平方与该两两输出对象的特征属性的乘积的比值,获得该两两输出对象的相似度;以及
根据该两两输出对象的相似度,获得所述相似度矩阵。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述输出单元,具体用于
对所述相似度矩阵进行SVD处理,以获得分解矩阵;
根据所述分解矩阵,获得所述每个输出对象的输出位置;以及
根据所述每个输出对象的输出位置,输出所述每个输出对象。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述输出单元,具体用于
根据所述相似度矩阵,计算两两输出对象的理论距离;
根据预先设置的至少一个特定种子,利用随机函数,获得所述每个输出对象的至少一个随机位置;
分别以所述每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行迭代处理,以获得所述每个输出对象的迭代位置;
根据所述每个输出对象的迭代位置,获得所述每个输出对象的输出位置;以及
根据所述每个输出对象的输出位置,输出所述每个输出对象。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述输出单元,具体用于
将1与两两输出对象的相似度的比值,作为该两两输出对象的距离;或者
若两两输出对象的相似度不为0,将1与该两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;若两两输出对象的相似度为0,将1与最小的两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;N为大于0且小于1的数。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,N为0.5。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述输出单元,具体用于
根据所述相似度矩阵,在二维平面空间或三维立体空间上,对所述每个输出对象进行可视化输出。
本发明的另一方面,提供一种设备,包括:
一个或者多个处理器;
存储器;
一个或者多个程序,所述一个或者多个程序存储在所述存储器中,当被所述一个或者多个处理器执行时:
获取至少两个输出对象中每个输出对象的特征属性;
根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;
根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;
根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
本发明的另一方面,提供一种非易失性计算机存储介质,所述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:
获取至少两个输出对象中每个输出对象的特征属性;
根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;
根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;
根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
由上述技术方案可知,本发明实施例通过获取至少两个输出对象中每个输出对象的特征属性,进而根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集,以及根据所述每个输出对象的特征
属性和所述两两输出对象的特征属性的交集,获得相似度矩阵,使得能够根据所述相似度矩阵,对所述每个输出对象进行可视化输出,由于根据每个输出对象的特征属性和两两输出对象的特征属性的交集所获得的相似度矩阵,能够表示两两输出对象之间的双向关系,使得所获得的两两输出对象的相似度是两两输出对象真正的相似度,从而提高了输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,通过分别以每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行多次迭代处理,使得可以获得最优结果,能够进一步提高输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,预先设置特定种子,使得所进行的多次迭代处理中每个迭代处理的结果较为稳定,能够有效提高输出对象可视化的稳定性。
另外,采用本发明所提供的技术方案,通过改进两两输出对象的理论距离的计算方法,使得所输出的每个输出对象的输出位置间隔有所减小,能够进一步提高输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,能够有效提高用户体验。
为了更清楚地说明本发明实施例中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1为本发明一实施例提供的输出对象的数据处理方法的流程示意图;
图2为本发明另一实施例提供的输出对象的数据处理装置的结构示意图。
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的全部其他实施例,都属于本发明保护的范围。
需要说明的是,本发明实施例中所涉及的终端可以包括但不限于手机、个人数字助理(Personal Digital Assistant,PDA)、无线手持设备、平板电脑(Tablet Computer)、个人电脑(Personal Computer,PC)、MP3播放器、MP4播放器、可穿戴设备(例如,智能眼镜、智能手表、智能手环等)等。
另外,本文中术语“和/或”,仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“/”,一般表
示前后关联对象是一种“或”的关系。
图1为本发明一实施例提供的输出对象的数据处理方法的流程示意图,如图1所示。
101、获取至少两个输出对象中每个输出对象的特征属性。
所谓的输出对象,是指能够可视化输出的对象,例如,推广产品与其他推广产品之间的关系,或者推广产品与媒体属性、人口属性和地域属性之间的关系,等等。
所谓的推广产品,是指用以推广用户的信息的网络推广方式,可以包括但不限于搜索推广、网盟推广、品牌广告、百度联盟、广告管家、百度统计、百度商桥等,本实施例对此不进行特别限定。
102、根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集。
其中,两两输出对象的特征属性的交集,是指两个输出对象公共的特征属性。
103、根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵。
在一个具体的实现过程中,所述相似度矩阵中所包含的每个矩阵元素,可以代表一对输出对象的相似度,其主对角线上的矩阵元素,可以代表一个输出对象与自身的相似度。每个输出对象与自身的相似度可以预先指定,例如,0或1。
例如,相似度矩阵T可以为如下形式:
104、根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
需要说明的是,101~104的执行主体的部分或全部可以为位于本地终端的应用,或者还可以为设置在位于本地终端的应用中的插件或软件开发工具包(Software Development Kit,SDK)等功能单元,或者还可以为位于网络侧服务器中的处理引擎,或者还可以为位于网络侧的分布式系统,本实施例对此不进行特别限定。
可以理解的是,所述应用可以是安装在终端上的本地程序(nativeApp),或者还可以是终端上的浏览器的一个网页程序(webApp),本实施例对此不进行限定。
这样,通过获取至少两个输出对象中每个输出对象的特征属性,进而根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集,以及根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵,使得能够根据所述相似度矩阵,对所述每个输出对象进行可视化输出,由于根据每个输出对象的特征属性和两两输出对象的特征属性的交集所获得的相似度矩阵,能够表示两两输出对象之间的双向关系,使得所获得的两两输出对象的相似度是两两输出对象真正的相似度,从而提高了输出对象可视化的可靠性。
可选地,在本实施例的一个可能的实现方式中,所述输出对象的特征属性可以包括但不限于如下属性:
输出对象所属搜索关键词所对应的人数。
可选地,在本实施例的一个可能的实现方式中,在103中,具体可以根据两两输出对象的特征属性的交集的平方与该两两输出对象的特征属性的乘积的比值,获得该两两输出对象的相似度,进而,则可以根据该两两输出对象的相似度,获得所述相似度矩阵。
这种实现方式中,实际上可以理解为,将,两两输出对象的特征属性的交集分别与该两两输出对象中的每个输出对象的比值,相乘,得到该两两输出对象的相似度。再由两两输出对象的相似度,构成所述相似度矩阵。
这种根据每个输出对象的特征属性和两两输出对象的特征属性的交集所获得的相似度,能够表示两两输出对象之间的双向关系,这样,所获得的两两输出对象的相似度是两两输出对象真正的相似度,能够更加准确地表征出两两输出对象之间的关系。
可选地,在本实施例的一个可能的实现方式中,在104中,具体可以对所述相似度矩阵进行奇异值分解(Singular Value Decomposition,SVD)处理,以获得分解矩阵,进而,则可以根据所述分解矩阵,获得所述每个输出对象的输出位置。然后,则根据所述每个输出对象的输出位置,输出所述每个输出对象。
在一个具体的实现过程中,SVD处理的详细描述可以参见现有技术中的相关内容,此处不再赘述。
在另一个具体的实现过程中,除了根据所述分解矩阵,获得所述每个输出对象的输出位置即以指定坐标轴定义的输出空间中的坐标值,之外,还可以进一步根据所述分解矩阵,获得该输出空间中的所述指定坐
标轴的贡献度,以供输出给用户。
可选地,在本实施例的一个可能的实现方式中,在104中,具体可以根据所述相似度矩阵,计算两两输出对象的理论距离,进而,则可以根据预先设置的至少一个特定种子,利用随机函数,获得所述每个输出对象的至少一个随机位置。接着,分别以所述每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行迭代处理,以获得所述每个输出对象的迭代位置,进而根据所述每个输出对象的迭代位置,获得所述每个输出对象的输出位置。然后,则根据所述每个输出对象的输出位置,输出所述每个输出对象。
这样,通过分别以每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行多次迭代处理,使得可以获得最优结果,能够进一步提高输出对象可视化的可靠性。
在这个实现方式中,所谓的输出对象的随机位置,并不是一个真正的随机位置,而实际上是一个伪随机位置。获得这个伪随机位置,所依据的伪随机函数,是以一个种子随机函数作为初始条件,然后用一定的算法不停迭代所产生的。这样,预先设置特定种子,使得所进行的多次迭代处理中每个迭代处理的结果较为稳定,能够有效提高输出对象可视化的稳定性。
在一个具体的实现过程中,具体可以将1与两两输出对象的相似度的比值,作为该两两输出对象的距离。这种传统的距离定义方式,所获得的距离的精度非常差,尤其是当一个相似度矩阵中所包含的矩阵元素
的取值相差比较大时,所输出的每个输出对象的输出位置间隔会非常大,这样,会导致输出对象的可视化的可靠性的降低。
在另一个具体的实现过程中,具体可以若两两输出对象的相似度不为0,将1与该两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;若两两输出对象的相似度为0,将1与最小的两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;N为大于0且小于1的数。这种改进的距离定义方式,使得所输出的每个输出对象的输出位置间隔有所减小,能够进一步提高输出对象可视化的可靠性。
优选地,N可以取值为0.5。
可选地,在本实施例的一个可能的实现方式中,在104中,具体可以根据所述相似度矩阵,在二维平面空间或三维立体空间等输出空间上,对所述每个输出对象进行可视化输出。这样,通过将所述每个输出对象在输出空间上进行可视化输出,使得能够清晰地确定每个输出对象之间的关系。
本实施例中,通过获取至少两个输出对象中每个输出对象的特征属性,进而根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集,以及根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵,使得能够根据所述相似度矩阵,对所述每个输出对象进行可视化输出,由于根据每个输出对象的特征属性和两两输出对象的特征属性的交集所获得的相似度矩阵,能够表示两两输出对象之间的双向关系,使得所获得的两两输出对象的相似度是两两输出对象真正的相似度,从而提高了输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,通过分别以每个输出对象的
至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行多次迭代处理,使得可以获得最优结果,能够进一步提高输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,预先设置特定种子,使得所进行的多次迭代处理中每个迭代处理的结果较为稳定,能够有效提高输出对象可视化的稳定性。
另外,采用本发明所提供的技术方案,通过改进两两输出对象的理论距离的计算方法,使得所输出的每个输出对象的输出位置间隔有所减小,能够进一步提高输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,能够有效提高用户体验。
需要说明的是,对于前述的各方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本发明并不受所描述的动作顺序的限制,因为依据本发明,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定是本发明所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其他实施例的相关描述。
图2为本发明另一实施例提供的输出对象的数据处理装置的结构示意图,如图2所示。本实施例的输出对象的数据处理装置可以包括获取单元21、分析单元22和输出单元23。其中,获取单元21,用于获取至少两个输出对象中每个输出对象的特征属性;分析单元22,用于根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;
所述分析单元22,还用于根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;输出单元23,用于根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
需要说明的是,本实施例所提供的输出对象的数据处理装置的执行主体的部分或全部可以为位于本地终端的应用,或者还可以为设置在位于本地终端的应用中的插件或软件开发工具包(Software Development Kit,SDK)等功能单元,或者还可以为位于网络侧服务器中的处理引擎,或者还可以为位于网络侧的分布式系统,本实施例对此不进行特别限定。
可以理解的是,所述应用可以是安装在终端上的本地程序(nativeApp),或者还可以是终端上的浏览器的一个网页程序(webApp),本实施例对此不进行限定。
可选地,在本实施例的一个可能的实现方式中,所述输出对象的特征属性可以包括但不限于如下属性:
输出对象所属搜索关键词所对应的人数。
可选地,在本实施例的一个可能的实现方式中,所述分析单元22,具体可以用于根据两两输出对象的特征属性的交集的平方与该两两输出对象的特征属性的乘积的比值,获得该两两输出对象的相似度;以及根据该两两输出对象的相似度,获得所述相似度矩阵。
可选地,在本实施例的一个可能的实现方式中,所述输出单元23,具体可以用于对所述相似度矩阵进行SVD处理,以获得分解矩阵;根据所述分解矩阵,获得所述每个输出对象的输出位置;以及根据所述每个输出对象的输出位置,输出所述每个输出对象。
可选地,在本实施例的一个可能的实现方式中,所述输出单元23,
具体可以用于根据所述相似度矩阵,计算两两输出对象的理论距离;根据预先设置的至少一个特定种子,利用随机函数,获得所述每个输出对象的至少一个随机位置;分别以所述每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行迭代处理,以获得所述每个输出对象的迭代位置;根据所述每个输出对象的迭代位置,获得所述每个输出对象的输出位置;以及根据所述每个输出对象的输出位置,输出所述每个输出对象。
在一个具体的实现过程中,所述输出单元23,具体可以用于将1与两两输出对象的相似度的比值,作为该两两输出对象的距离。
在另一个具体的实现过程中,所述输出单元23,具体可以用于若两两输出对象的相似度不为0,将1与该两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;若两两输出对象的相似度为0,将1与最小的两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;N为大于0且小于1的数。
优选地,N可以取值为0.5。
可选地,在本实施例的一个可能的实现方式中,所述输出单元23,具体可以用于根据所述相似度矩阵,在二维平面空间或三维立体空间上,对所述每个输出对象进行可视化输出。
需要说明的是,图1对应的实施例中方法,可以由本实施例提供的输出对象的数据处理装置实现。详细描述可以参见图1对应的实施例中的相关内容,此处不再赘述。
本实施例中,通过获取单元获取至少两个输出对象中每个输出对象的特征属性,进而由分析单元根据所述每个输出对象的特征属性,获得
两两输出对象的特征属性的交集,以及根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵,使得输出单元能够根据所述相似度矩阵,对所述每个输出对象进行可视化输出,由于根据每个输出对象的特征属性和两两输出对象的特征属性的交集所获得的相似度矩阵,能够表示两两输出对象之间的双向关系,使得所获得的两两输出对象的相似度是两两输出对象真正的相似度,从而提高了输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,通过分别以每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行多次迭代处理,使得可以获得最优结果,能够进一步提高输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,预先设置特定种子,使得所进行的多次迭代处理中每个迭代处理的结果较为稳定,能够有效提高输出对象可视化的稳定性。
另外,采用本发明所提供的技术方案,通过改进两两输出对象的理论距离的计算方法,使得所输出的每个输出对象的输出位置间隔有所减小,能够进一步提高输出对象可视化的可靠性。
另外,采用本发明所提供的技术方案,能够有效提高用户体验。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本发明所提供的几个实施例中,应该理解到,所揭露的系统,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例
仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元,可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)或处理器(processor)执行本发明各个实施例所述方法的部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是:以上实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述实施例对本发明进行了详细的说明,本领域的
普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。
Claims (18)
- 一种输出对象的数据处理方法,其特征在于,包括:获取至少两个输出对象中每个输出对象的特征属性;根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
- 根据权利要求1所述的方法,其特征在于,所述输出对象的特征属性包括:输出对象所属搜索关键词所对应的人数。
- 根据权利要求1所述的方法,其特征在于,所述根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵,包括:根据两两输出对象的特征属性的交集的平方与该两两输出对象的特征属性的乘积的比值,获得该两两输出对象的相似度;根据该两两输出对象的相似度,获得所述相似度矩阵。
- 根据权利要求1所述的方法,其特征在于,所述根据所述相似度矩阵,对所述每个输出对象进行可视化输出,包括:对所述相似度矩阵进行SVD处理,以获得分解矩阵;根据所述分解矩阵,获得所述每个输出对象的输出位置;根据所述每个输出对象的输出位置,输出所述每个输出对象。
- 根据权利要求1所述的方法,其特征在于,所述根据所述相似度 矩阵,对所述每个输出对象进行可视化输出,包括:根据所述相似度矩阵,计算两两输出对象的理论距离;根据预先设置的至少一个特定种子,利用随机函数,获得所述每个输出对象的至少一个随机位置;分别以所述每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行迭代处理,以获得所述每个输出对象的迭代位置;根据所述每个输出对象的迭代位置,获得所述每个输出对象的输出位置;根据所述每个输出对象的输出位置,输出所述每个输出对象。
- 根据权利要求5所述的方法,其特征在于,所述根据所述相似度矩阵,计算两两输出对象的距离,包括:将1与两两输出对象的相似度的比值,作为该两两输出对象的距离;或者若两两输出对象的相似度不为0,将1与该两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;若两两输出对象的相似度为0,将1与最小的两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;N为大于0且小于1的数。
- 根据权利要求6所述的方法,其特征在于,N为0.5。
- 根据权利要求1~7任一权利要求所述的方法,其特征在于,所述根据所述相似度矩阵,对所述每个输出对象进行可视化输出,包括:根据所述相似度矩阵,在二维平面空间或三维立体空间上,对所述每个输出对象进行可视化输出。
- 一种输出对象的数据处理装置,其特征在于,包括:获取单元,用于获取至少两个输出对象中每个输出对象的特征属性;分析单元,用于根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;所述分析单元,还用于根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;输出单元,用于根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
- 根据权利要求9所述的装置,其特征在于,所述输出对象的特征属性包括:输出对象所属搜索关键词所对应的人数。
- 根据权利要求9所述的装置,其特征在于,所述分析单元,具体用于根据两两输出对象的特征属性的交集的平方与该两两输出对象的特征属性的乘积的比值,获得该两两输出对象的相似度;以及根据该两两输出对象的相似度,获得所述相似度矩阵。
- 根据权利要求9所述的装置,其特征在于,所述输出单元,具体用于对所述相似度矩阵进行SVD处理,以获得分解矩阵;根据所述分解矩阵,获得所述每个输出对象的输出位置;以及根据所述每个输出对象的输出位置,输出所述每个输出对象。
- 根据权利要求9所述的装置,其特征在于,所述输出单元,具体用于根据所述相似度矩阵,计算两两输出对象的理论距离;根据预先设置的至少一个特定种子,利用随机函数,获得所述每个输出对象的至少一个随机位置;分别以所述每个输出对象的至少一个随机位置中每个随机位置为该输出对象的初始位置,根据所述两两输出对象的理论距离,进行迭代处理,以获得所述每个输出对象的迭代位置;根据所述每个输出对象的迭代位置,获得所述每个输出对象的输出位置;以及根据所述每个输出对象的输出位置,输出所述每个输出对象。
- 根据权利要求13所述的装置,其特征在于,所述输出单元,具体用于将1与两两输出对象的相似度的比值,作为该两两输出对象的距离;或者若两两输出对象的相似度不为0,将1与该两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;若两两输出对象的相似度为0,将1与最小的两两输出对象的相似度的比值的N次方,作为该两两输出对象的距离;N为大于0且小于1的数。
- 根据权利要求14所述的装置,其特征在于,N为0.5。
- 根据权利要求9~15任一权利要求所述的装置,其特征在于,所述输出单元,具体用于根据所述相似度矩阵,在二维平面空间或三维立体空间上,对所述每个输出对象进行可视化输出。
- 一种设备,包括:一个或者多个处理器;存储器;一个或者多个程序,所述一个或者多个程序存储在所述存储器中,当被所述一个或者多个处理器执行时:获取至少两个输出对象中每个输出对象的特征属性;根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
- 一种非易失性计算机存储介质,所述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:获取至少两个输出对象中每个输出对象的特征属性;根据所述每个输出对象的特征属性,获得两两输出对象的特征属性的交集;根据所述每个输出对象的特征属性和所述两两输出对象的特征属性的交集,获得相似度矩阵;根据所述相似度矩阵,对所述每个输出对象进行可视化输出。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510349469.0 | 2015-06-23 | ||
| CN201510349469.0A CN104951545B (zh) | 2015-06-23 | 2015-06-23 | 输出对象的数据处理方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016206188A1 true WO2016206188A1 (zh) | 2016-12-29 |
Family
ID=54166203
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/088391 Ceased WO2016206188A1 (zh) | 2015-06-23 | 2015-08-28 | 输出对象的数据处理方法、装置、设备及非易失性计算机存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104951545B (zh) |
| WO (1) | WO2016206188A1 (zh) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299186A (zh) * | 2018-11-05 | 2019-02-01 | 北京锐安科技有限公司 | 数据处理方法、装置、设备和存储介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080285861A1 (en) * | 2007-05-16 | 2008-11-20 | Canon Kabushiki Kaisha | Information processing method and information processing apparatus |
| CN102902821A (zh) * | 2012-11-01 | 2013-01-30 | 北京邮电大学 | 基于网络热点话题的图像高级语义标注、检索方法及装置 |
| CN103246685A (zh) * | 2012-02-14 | 2013-08-14 | 株式会社理光 | 将对象实例的属性规则化为特征的方法和设备 |
| CN104572825A (zh) * | 2014-12-04 | 2015-04-29 | 百度在线网络技术(北京)有限公司 | 信息的推荐方法和装置 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708180B (zh) * | 2012-05-09 | 2014-02-26 | 北京华电天仁电力控制技术有限公司 | 基于实时历史库的机组运行模式数据挖掘方法 |
| CN103500208B (zh) * | 2013-09-30 | 2016-08-17 | 中国科学院自动化研究所 | 结合知识库的深层数据处理方法和系统 |
| CN104318340B (zh) * | 2014-09-25 | 2017-07-07 | 中国科学院软件研究所 | 基于文本履历信息的信息可视化方法及智能可视分析系统 |
-
2015
- 2015-06-23 CN CN201510349469.0A patent/CN104951545B/zh active Active
- 2015-08-28 WO PCT/CN2015/088391 patent/WO2016206188A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080285861A1 (en) * | 2007-05-16 | 2008-11-20 | Canon Kabushiki Kaisha | Information processing method and information processing apparatus |
| CN103246685A (zh) * | 2012-02-14 | 2013-08-14 | 株式会社理光 | 将对象实例的属性规则化为特征的方法和设备 |
| CN102902821A (zh) * | 2012-11-01 | 2013-01-30 | 北京邮电大学 | 基于网络热点话题的图像高级语义标注、检索方法及装置 |
| CN104572825A (zh) * | 2014-12-04 | 2015-04-29 | 百度在线网络技术(北京)有限公司 | 信息的推荐方法和装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104951545A (zh) | 2015-09-30 |
| CN104951545B (zh) | 2018-07-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11238065B1 (en) | Systems and methods for generating and implementing knowledge graphs for knowledge representation and analysis | |
| JP6759844B2 (ja) | 画像を施設に対して関連付けるシステム、方法、プログラム及び装置 | |
| CN107480158B (zh) | 基于相似性得分评估内容项目与图像的匹配的方法和系统 | |
| CN112396456B (zh) | 广告推送方法、装置、存储介质以及终端 | |
| CN105512180B (zh) | 一种搜索推荐方法及装置 | |
| CN107451832B (zh) | 推送信息的方法和装置 | |
| WO2018120424A1 (zh) | 基于位置服务的人群分类方法、装置、设备和存储介质 | |
| CN113434755A (zh) | 页面的生成方法、装置、电子设备及存储介质 | |
| CN111279334A (zh) | 利用上下文分析的搜索查询增强 | |
| US20150302515A1 (en) | Method, apparatus, and system for simulating objects | |
| CN107870984A (zh) | 识别搜索词的意图的方法和装置 | |
| US10915586B2 (en) | Search engine for identifying analogies | |
| CN106528611A (zh) | 一种基于互联网点评数据的分析方法 | |
| CN113688310A (zh) | 一种内容推荐方法、装置、设备及存储介质 | |
| CN110807095A (zh) | 一种物品匹配方法和装置 | |
| WO2016107455A1 (zh) | 一种信息匹配处理方法和装置 | |
| CN113722593A (zh) | 事件数据处理方法、装置、电子设备和介质 | |
| JP2019520617A (ja) | 広告生成方法、コンピュータ読み取り可能な記憶媒体及びシステム | |
| US20140149465A1 (en) | Feature rich view of an entity subgraph | |
| US20220398244A1 (en) | Query method and device and storage medium | |
| Jowaheer et al. | A BINAR (1) time-series model with cross-correlated COM–Poisson innovations | |
| CN111767420A (zh) | 一种生成服饰搭配数据的方法和装置 | |
| CN113779186A (zh) | 一种文本生成方法和装置 | |
| Alkan et al. | A comparison of different procedures for principal component analysis in the presence of outliers | |
| WO2015148420A1 (en) | User inactivity aware recommendation system |
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: 15896083 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 15896083 Country of ref document: EP Kind code of ref document: A1 |
