Method and device for updating knowledge graph

[0032] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0033] The embodiment of the present invention provides a method for updating the knowledge map, such as figure 2 As shown, the method mainly includes:

[0034] 101. Set a change period for the attribute of the entity in the data layer of the knowledge map.

[0035] Specifically, the knowledge graph is mainly divided into a schema layer and a data layer, and the data layer is a specific instance of the schema layer. In practical applications, each attribute value in the data layer may change dynamically or may be fixed. In order to quickly know which attribute value needs to be updated, it is necessary to know the change period of the attribute first, and then determine the next update time according to the change period, and finally update the attribute value of the corresponding attribute at the next update time point.

[0036] When setting the change period for the attribute of the entity in the data layer of the knowledge graph, the specific solutions that can be adopted include but are not limited to the following two:

[0037] (1) It is directly set by professionals according to the data layer.

[0038] Specifically, the knowledge map update device can first extract the entities to be set and the attributes of the entities from the data layer of the knowledge map, and then professionals can set the change cycle for the attributes of each entity according to the correspondence between entities and attributes .

[0039] (2) Professionals first set it according to the pattern layer, and then the knowledge map update device automatically sets the change cycle for the attributes in the data layer according to the change cycle set for the pattern layer.

[0040] Specifically, the knowledge graph update device may first extract the attributes of entities in the schema layer of the knowledge graph; then receive the change period set for each attribute; finally, according to the change cycle set for each attribute in the schema layer, provide The attribute setting change cycle of each entity in the above data layer. That is to say, the knowledge map update device can first extract the entities to be set and the attributes of the entities from the schema layer of the knowledge map, and then professionals can set the Change cycle. Finally, the knowledge map update device searches for the corresponding attribute in the data layer according to the pattern layer and the change cycle set for each attribute in the pattern layer, and sets the change cycle for the corresponding attribute.

[0041] For example, if in the schema layer, "person" has an attribute "age", and the change period set for this attribute is "1 year", then in the data layer, the specific instance "Zhang San" corresponding to "person", "Li Si", "Wang Wu" and so on, the change cycle of the corresponding attribute "age" is "1 year".

[0042] Since the data layer is a specific instance of the schema layer, and in practical applications, there are often thousands or even more specific instances of the same schema, the data volume of the data layer is much larger than that of the schema layer. Therefore, the time spent manually setting the change cycle for all attributes in the schema layer is far less than the time spent manually setting the change cycle for all attributes in the data layer, so the second method can greatly improve the time spent on setting the change cycle for all attributes in the data layer. Efficiency of the change cycle.

[0043]It should be noted that in practical applications, when the user only cares about whether the relevant information of a certain entity is the latest information, it can only be set for this entity; when the user only cares about whether the relevant information of certain entities is the latest information Yes, it can be set for multiple entities; when the user cares whether the related information of all entities is up-to-date or when the entities related to different users are different, it can be set for all entities. Therefore, this step may specifically be: firstly determine at least one entity to be set in the data layer of the knowledge graph, and then set a change cycle for the attribute of the determined entity.

[0044] Wherein, a specific implementation manner of determining at least one entity to be set in the data layer of the knowledge graph may be: receiving at least one entity to be set in the data layer of the knowledge graph input by a user.

[0045] 102. Based on the current time and the change period of the attribute, determine the next update time of the attribute value corresponding to the attribute.

[0046] After obtaining the change cycle of each attribute, in order to update the attribute value that needs to be updated in a more timely manner, the next update time of each attribute value can be obtained by calculating the sum of the current time and each change cycle separately, so as to reach the next time When the time is updated, the corresponding attribute value is updated.

[0047] Exemplarily, if the current time is 2016-07-15 00:00:00, and the change period is 1 day, the next update time is 2016-07-16 00:00:00.

[0048] 103. Update the corresponding attribute value in the data layer according to the next update time of the attribute value.

[0049] After calculating the next update time of each attribute value in the data layer, the next update time of each attribute value can be scanned periodically or in real time. When the current scan time is equal to a certain next update time, the knowledge map update device can Determine that the update time of the attribute value corresponding to the next update time is up, and immediately find the corresponding crawler parameters to crawl the network data, and replace the latest attribute value in the crawled network data with the corresponding attribute value in the data layer to realize the attribute The update operation for the value. That is to say, if it is determined that the next update time of a certain attribute value is reached, the corresponding attribute value in the data layer may be updated by crawling network data. Among them, when it is determined that the next update of a certain attribute value is reached, the corresponding attribute value in the data layer can be updated by crawling network data immediately, or within a period of time after the next update time of a certain attribute value is determined. , update the corresponding attribute value in the data layer by crawling network data, as long as it is known that a certain attribute value may have just been updated in the web page, it can be updated within a relatively short period of time.

[0050] Wherein, the crawler parameters mainly include: the URL (Uniform Resource Locator, uniform resource locator) to be crawled and the location information of the information to be crawled in the DOM (Document Object Model, Document Object Model) tree of the web page.

[0051] The method for updating the knowledge map provided by the embodiment of the present invention can first set the change cycle for the attribute of the entity in the knowledge map data layer, and then determine the next update of the attribute value corresponding to the attribute based on the current time and the change cycle set for the attribute Time, and finally through time monitoring to determine which attribute values ​​​​reach their own next update time, and when it is determined that a certain attribute value has reached its own corresponding next update time, immediately update the corresponding in the data layer by crawling network data Compared with regular full update, it not only can update the attribute value that needs to be updated in time, but also does not need to spend time and resources to update other attribute values ​​that have not reached its corresponding next update time, which not only enhances the ability to update the knowledge graph The timeliness also improves the efficiency of updating the knowledge map.

[0052] Further, according to figure 2 As shown in the method, another embodiment of the present invention also provides a method for updating the knowledge map, such as image 3 As shown, the method mainly includes:

[0053] 201. Set a change cycle for attributes of entities in the data layer of the knowledge graph.

[0054] The specific implementation manner of this step is consistent with the specific implementation manner of the above-mentioned step 101, and will not be repeated here.

[0055] 202. Based on the current time and the change period of the attribute, determine the next update time of the attribute value corresponding to the attribute.

[0056] The specific implementation manner of this step is consistent with the specific implementation manner of the above-mentioned step 102, and will not be repeated here.

[0057] 203. Update the corresponding attribute value in the data layer according to the next update time of the attribute value.

[0058] When the update time arrives, in order to quickly obtain the relevant information of the attribute value that needs to be updated, and then quickly update the knowledge map, this step can be specifically refined into the following steps (a)-(c):

[0059] (a) Build an index database.

[0060] Wherein, the index database mainly includes the correspondence between entities in the data layer, attributes of entities, change periods of attributes, next update time, and crawler parameters required for crawling network data. In addition, when building an index database, the next update time can be set as the primary key, so as to quickly find all content under the same next update time according to the primary key.

[0061] The embodiment of the present invention does not limit the type of the index database, for example, it may be a MySQL type or an Hbase type.

[0062] Exemplarily, the storage format of the index database may be as shown in Table 1.

[0063] Table 1

[0064]

[0065] (b) Scanning the index database to filter out the next update time that is the same as the current scanning time.

[0066] After the index database is established, the index database can be scanned in real time or periodically to determine whether the index database contains the next update time that is the same as the current scan time. When it is determined that the next update time is the same as the current scan time, the next The first update time and the information corresponding to the next update time are filtered out, and the filtered information is stored in the memory storage queue, so that the corresponding attribute values ​​in the knowledge map can be updated based on these information. Wherein, the specific type of the memory storage queue is not limited, for example, it may be a metaq memory storage queue.

[0067] For example, if table 1 is scanned every minute, the scan result at 2016-07-16 00:00:00 is table 2, and the scan result at 2016-07-16 15:05:00 is table 3. The scanning result at 2017-07-16 01:00:00 is Table 4.

[0068] Table 2

[0069]

[0070] table 3

[0071]

[0072] Table 4

[0073]

[0074] (c) According to the selected crawler parameters corresponding to the next update time, crawl the attribute value of the corresponding attribute, and replace the crawled attribute value with the corresponding attribute value in the data layer.

[0075] When filtering out the next update time that is the same as the current scan time and the information corresponding to the next update time, the corresponding attributes can be crawled according to the crawler parameters (including URL and DOM tree location information) corresponding to each next update time , and then replace the crawled attribute values ​​with the corresponding attribute values ​​in the knowledge map data layer, so as to realize the update of the knowledge map. Among them, after the knowledge map update device crawls the attribute values ​​from the network according to the crawler parameters, it can store these attribute values ​​in the preset storage space, so that when the attribute value is updated in the future, it can be quickly searched from the preset storage space. The desired attribute value.

[0076] The specific implementation of replacing the crawled attribute value with the corresponding attribute value in the data layer may be as follows: the knowledge map update device first selects the entity corresponding to the next update time in the index database, and the attribute of the entity , searching for the corresponding attribute value in the data layer of the knowledge graph; and then replacing the found attribute value with the crawled attribute value.

[0077] Wherein, according to the entity and the attribute of the entity corresponding to the selected next update time in the index database, the specific implementation method of searching for the corresponding attribute value in the data layer of the knowledge graph is as follows: first, according to the selected entity Find the corresponding entity in the data layer of the indicated image, and then find the required attribute in the data layer according to the filtered attribute and the found entity, and finally find the attribute value according to the found attribute.

[0078] Further, when the index database also includes attribute values, the crawled attribute values ​​can be replaced with the corresponding attribute values ​​in the index database, so that when the attribute values ​​need to be queried from the index database, the latest attribute value.

[0079] 204. Based on the change period of the attribute, update the next update time corresponding to the attribute.

[0080] After the update of the knowledge graph is completed, the next update time needs to be updated, so as to realize the next update of the knowledge graph according to the updated next update time.

[0081] In practical applications, in addition to periodically changing attributes and fixed attributes, there is also an attribute with an unknown change period. For example, a person's place of residence may change during this lifetime, but it is not known when the change will occur, or it may not change at all, so figure 1 The change cycle of an attribute "lived in" of the entity "Li Lei" is unknown.

[0082] When the knowledge graph that needs to be updated contains attributes whose change period is dynamic, you can add a period feature to indicate whether the change period will change dynamically and the number of updates of the change period when building the index database, so that the follow-up can be based on these two Information to update the next update time.

[0083] The specific implementation of this step will be described in detail below for the two situations of fixed change period and dynamic change period:

[0084] (1) If the period feature is fixed, update the corresponding next update time based on the current change period corresponding to the period feature.

[0085] For example, the change period of the attribute "weather of the day" of the entity "Shanghai" is fixed, and the change period is 1 day. Therefore, after updating the corresponding attribute value in the knowledge graph, the next update time is still the current time +1 sky.

[0086] (2) If the periodic feature is a dynamic change, update the change period corresponding to the periodic feature based on the preset periodic update algorithm and the update times corresponding to the periodic feature, and update the corresponding change period based on the updated change period. Next update time.

[0087]Since we don't know the change period of some attributes (such as "living in"), for different instances, we can first set the change period to the minimum value (such as 1 day), and then try it out by increasing the change period. Find a more suitable change period from it. In the specific trial process, since the starting value of the change period is relatively small, it can be more frequent when starting the trial; when the data has not changed after multiple reptiles are found, it can be considered that the attribute is relatively stable for this instance , so the change period can be enlarged, so that the trial becomes less and less until the change period is stable (that is, reaches the maximum value set manually); and if the data changes after a crawler is found, the change period can be Reset to the minimum value and restart the polling heuristic, so that personalized updates to different instances of the same attribute can be achieved. Among them, the purpose of setting the maximum value for the change period is to effectively control the update frequency, and to balance the timeliness of the update and the resources consumed by the update.

[0088] It can be seen that in the process of testing the change cycle, the change cycle needs to be made larger and larger. In order to make the change cycle larger and larger, the preset cycle update algorithm can be set as an incremental function, and the change cycle increases with the number of updates. Increase and increase, so that the change period can be increased by increasing the update times; the preset period update algorithm can also be set as a decreasing function, and the change period increases with the decrease of the update times, so that The change period is increased by reducing the number of updates.

[0089] From the above analysis, it can be seen that the specific implementation of updating the change cycle can be the following steps (A)-(B):

[0090] (A) The knowledge map updating device adjusts the number of updates corresponding to the periodic feature according to the difference between the crawled attribute value corresponding to the periodic feature and the original attribute value corresponding to the periodic feature.

[0091] Specifically, when the preset cycle update algorithm is an incremental function, and the change cycle increases with the increase of the update times, the specific implementation of this step can be: when the crawled attribute corresponding to the cycle feature When the value is the same as the original attribute value corresponding to the periodic feature, the knowledge map updating device can increase the number of updates corresponding to the periodic feature by a preset number of times; when the crawled attribute value corresponding to the periodic feature is the same as the periodic When the original attribute values ​​corresponding to the features are different, the knowledge graph update device sets the update times corresponding to the periodic features to zero.

[0092] That is to say, when using the current change cycle to update the attribute value in the knowledge map, if the attribute value crawled from the network side is the same as the original attribute value, it means that the real update time has not yet arrived, so you can update by adjusting the The number of times to increase the change period, and check whether the attribute value changes again. When an attribute value crawled from the network side is different from the original attribute value, it is determined that the attribute value has been updated, and for attribute values ​​that are not frequently updated, if a sudden update occurs, it is likely that It will also be updated, so the change period can be adjusted to the minimum value by adjusting the number of updates to zero, so that the property value can be updated in time next time.

[0093] Among them, the original attribute value corresponding to the periodic feature can be obtained from the knowledge map according to the entity in the index database, the attribute of the entity, or directly from the index data (that is, the index database can record specific attributes value). When the index database also includes periodic features, update times, and attribute values, the specific form of the index database can be shown in Table 5.

[0094] table 5

[0095]

[0096]

[0097] It should be noted that the preset number of times in this step may be 1 or other values, depending on the specific situation.

[0098] In addition, when the update algorithm of the preset period is a decreasing function, and the change period increases as the number of updates decreases, the specific implementation of this step can be as follows: if the crawled attribute value corresponding to the period feature The original attribute value corresponding to the periodic feature is the same, then the number of updates corresponding to the periodic feature is subtracted from the preset number of times; if the crawled attribute value corresponding to the periodic feature is the same as the original attribute value corresponding to the periodic feature If the attribute values ​​are different, set the number of updates corresponding to the period feature to a preset maximum number of times, where the preset maximum number of times is the number of times that can make the change period the minimum value in the preset period update algorithm. Wherein, the principle of this process is similar to the principle of the above-mentioned incremental function, and will not be repeated here.

[0099] (B) Using the adjusted number of updates as an input parameter of the preset period update algorithm to calculate the updated change period.

[0100] Among them, when the preset update algorithm is an incremental function, in practical applications, Among them, t represents the number of updates, and t>=0, day max is the maximum value of the change period set.

[0101] Specifically, the above formula increases monotonically when t>=0; when t=0, the change period=1, and when t=+∞, the change period takes the maximum value day max; and when t increases from 0 to t1, the change period increases slowly, when t increases from t1 to t2, the change period increases faster, when t increases from t2 to t3, the change period begins to slow down again, and finally t tends to + When ∞, converge to day max. Therefore, this formula is very consistent with people's thinking process of exploring the cycle of change.

[0102] The method for updating the knowledge map provided by the embodiment of the present invention can not only set the change period for the attribute of the entity in the knowledge map data layer, and calculate the next update time of the attribute value, but also can obtain the change period, the next update time, Other basic information about attributes is stored in the index database, so that the next update time is used as the primary key to quickly scan out all relevant information about the attribute value that needs to be updated, and quickly update the attribute value in the knowledge map based on this information, thereby further improving the update. The efficiency of knowledge graph. In addition, when this update is completed and the next time monitoring is started by updating the next update time, the embodiment of the present invention also updates the next update time according to the periodic characteristics of the change period, that is, for a fixed change period, the embodiment of the present invention directly updates the corresponding next update time according to the change period, and for the dynamically changing change period, the embodiment of the present invention uses a preset algorithm to dynamically calculate the next update time, so that each attribute The next update time of the value is as close as possible to the next update time of the attribute value, which further enhances the timeliness of updating the knowledge graph.

[0103] Further, according to the above method embodiment, another embodiment of the present invention also provides a device for updating the knowledge map, such as Figure 4 As shown, the device mainly includes: a setting unit 31 , a determining unit 32 and an updating unit 33 . in,

[0104] The setting unit 31 is used to set the change period for the attribute of the entity in the data layer of the knowledge map;

[0105] A determining unit 32, configured to determine the next update time of the attribute value corresponding to the attribute based on the current time and the change period of the attribute;

[0106] The update unit 33 is configured to update the corresponding attribute value in the data layer according to the next update time of the attribute value.

[0107] optional, such as Figure 5 As shown, the setting unit 31 includes:

[0108] An extraction module 311, configured to extract attributes of entities in the schema layer of the knowledge graph;

[0109] A receiving module 312, configured to receive a change period set for the attribute extracted by the extraction module 311;

[0110] The setting module 313 is configured to set a change cycle for the attribute in the data layer according to the change cycle set for the attribute in the schema layer received by the receiving module 312 .

[0111] Optionally, the updating unit 33 is configured to update the corresponding attribute value in the data layer by crawling network data when it is determined that the next update time of the attribute value is reached.

[0112] optional, such as Figure 5 As shown, the update unit 33 includes:

[0113] Establishment module 331, used to establish an index database, the index database includes the correspondence between entities in the data layer, entity attributes, attribute change periods, next update time, and crawler parameters required for crawling network data relation;

[0114] A scanning and screening module 332, configured to scan the index database established by the establishment module, and filter out the next update time that is the same as the current scanning time;

[0115] The crawling module 333 is used to crawl the attribute value of the corresponding attribute according to the crawler parameter corresponding to the next update time screened out by the scanning and screening module 332;

[0116] The replacement module 334 is configured to replace the attribute value crawled by the crawling module 333 with the corresponding attribute value in the data layer.

[0117] optional, such as Figure 5 As shown, the replacement module 334 includes:

[0118] The search sub-module 3341 is used to search for the corresponding attribute value in the data layer of the knowledge map according to the entity and the attribute of the entity corresponding to the next update time screened out in the index database;

[0119] The replacement submodule 3342 is configured to replace the attribute value found by the search submodule with the crawled attribute value.

[0120] Optionally, the replacing module 334 is further configured to replace the corresponding attribute value in the index database with the attribute value crawled by the crawling module when the index database also includes attribute values.

[0121] Optionally, the updating unit 33 is further configured to update the next time corresponding to the attribute based on the change period of the attribute after updating the corresponding attribute value in the data layer according to the next update time of the attribute value. update time.

[0122] optional, such as Figure 5 As shown, the update unit 33 includes:

[0123] The first update module 335 is configured to: when the index database also includes a period characteristic indicating whether the change period will change dynamically and the number of updates of the change period, if the period characteristic is fixed, then based on the The current change period corresponding to the periodic feature, and the corresponding next update time for updating;

[0124] The second update module 336 is configured to update the change period corresponding to the periodic feature based on the preset periodic update algorithm and the update times corresponding to the periodic feature when the periodic feature changes dynamically, wherein the preset The periodic update algorithm is an incremental function, and the change period increases as the number of updates increases;

[0125] The third update module 337 is configured to update the corresponding next update time based on the updated change cycle obtained by the second update module 336 .

[0126] optional, such as Figure 5 As shown, the second update module 336 includes:

[0127] The adjustment sub-module 3361 is configured to adjust the number of updates corresponding to the periodic feature according to the difference between the crawled attribute value corresponding to the periodic feature and the original attribute value corresponding to the periodic feature;

[0128] The calculation sub-module 3362 is configured to use the number of updates adjusted by the adjustment sub-module 3361 as an input parameter of the preset period update algorithm to calculate the updated change period.

[0129] Optionally, the update algorithm for the preset period is an incremental function, and the change period increases as the number of updates increases;

[0130] Alternatively, the update algorithm for the preset period is a decreasing function, and the change period increases as the number of updates decreases.

[0131] Optionally, the adjustment sub-module 3361 is used to update the algorithm in the preset period as an incremental function, and the change period increases as the number of updates increases, when the crawling period corresponding to the period feature When the attribute value corresponding to the periodic feature is the same as the original attribute value corresponding to the periodic feature, increase the number of updates corresponding to the periodic feature by a preset number of times; when the crawled attribute value corresponding to the periodic feature is the same as the corresponding When the original attribute values ​​are different, the number of updates corresponding to the periodic feature is set to zero.

[0132] Among them, the preset periodic update algorithm that satisfies the above characteristics can be:

[0133]

[0134] Among them, t represents the number of updates, and t>=0, day max is the maximum value of the change period set.

[0135] Optionally, the adjustment sub-module 3361 is used to update the algorithm in the preset period as a decreasing function, and the change period increases as the number of updates decreases, if the crawling period corresponding to the period feature The attribute value of the attribute value corresponding to the periodic feature is the same as the original attribute value corresponding to the periodic feature, then the number of updates corresponding to the periodic feature is subtracted from the preset number of times; if the crawled attribute value corresponding to the periodic feature corresponds to the periodic feature different original attribute values, set the number of updates corresponding to the period feature to a preset maximum number of times, where the preset maximum number of times is the number of times that can make the change period the minimum value in the preset period update algorithm.

[0136] Optionally, the setting unit 31 is configured to determine at least one entity to be set in the data layer of the knowledge graph; and set a change period for the attribute of the determined entity.

[0137] The device for updating the knowledge map provided by the embodiment of the present invention can first set the change cycle for the attribute of the entity in the knowledge map data layer, and then calculate the next update of the attribute value corresponding to the attribute based on the current time and the change cycle set for the attribute Time, and finally through time monitoring to determine which attribute values ​​​​reach their own next update time, and when it is determined that a certain attribute value has reached its own corresponding next update time, immediately update the corresponding in the data layer by crawling network data Compared with regular full update, it not only can update the attribute value that needs to be updated in time, but also does not need to spend time and resources to update other attribute values ​​that have not reached its corresponding next update time, which not only enhances the ability to update the knowledge graph The timeliness also improves the efficiency of updating the knowledge map.

[0138] Further, according to the above-mentioned embodiment, another embodiment of the present invention also provides a storage medium, the storage medium stores a plurality of instructions, and the instructions are suitable for being loaded and executed by a processor:

[0139] Set the change period for the attribute of the entity in the data layer of the knowledge map;

[0140] Based on the current time and the change period of the attribute, determine the next update time of the attribute value corresponding to the attribute;

[0141] Update the corresponding attribute value in the data layer according to the next update time of the attribute value.

[0142] It should be noted that the instructions stored in the storage medium also include image 3 The content in the illustrated embodiment will not be repeated here.

[0143] Further, according to the above embodiments, another embodiment of the present invention also provides an electronic device, where the electronic device includes a storage medium and a processor;

[0144] The storage medium is used to store instructions executed by the processor and data required by the processor during the execution of the instructions;

[0145] The processor is configured to execute the following instructions:

[0146] Set the change period for the attributes in the data layer of the knowledge map;

[0147] Based on the current time and the change period of the attribute, determine the next update time of the attribute value corresponding to the attribute;

[0148] Update the corresponding attribute value in the data layer according to the next update time of the attribute value.

[0149] It should be noted that the instructions stored in the storage medium and the instructions executed by the processor also include image 3 The content in the illustrated embodiment will not be repeated here.

[0150] In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0151] It can be understood that related features in the above methods and devices can refer to each other. In addition, "first", "second" and so on in the above embodiments are used to distinguish each embodiment, and do not represent the advantages and disadvantages of each embodiment.

[0152] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above may refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0153] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with teaching based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not directed to any particular programming language. It is to be understood that various programming languages ​​may be used to implement the inventions described herein, and that the descriptions of specific languages ​​above are intended to disclose the best mode for carrying out the invention.

[0154] In the description provided herein, numerous specific details are set forth. It will be understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

[0155] Similarly, it is to be understood that in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or its description. This disclosure, however, should not be construed as reflecting an intention that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

[0156] Those skilled in the art will understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. Modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method so disclosed may be employed in any combination, unless at least some of such features and/or procedures or elements are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

[0157] Furthermore, those skilled in the art will appreciate that although some of the embodiments described herein include certain features, but not others, included in other embodiments, that combinations of features of different embodiments are intended to be within the scope of the invention within and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

[0158] Various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all functions of some or all of the components in the method and device for updating a knowledge graph according to an embodiment of the present invention . The present invention can also be implemented as apparatus or apparatus programs (eg, computer programs and computer program products) for performing part or all of the methods described herein. Such a program implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such signals may be downloaded from Internet sites, or provided on carrier signals, or in any other form.

[0159] It should be noted that the above-described embodiments illustrate rather than limit the invention, and that alternative embodiments may be devised by those skilled in the art without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. do not denote any order. These words can be interpreted as names.