CN103345489A - Event inquiry demand processing method and device - Google Patents
Event inquiry demand processing method and device Download PDFInfo
- Publication number
- CN103345489A CN103345489A CN2013102558766A CN201310255876A CN103345489A CN 103345489 A CN103345489 A CN 103345489A CN 2013102558766 A CN2013102558766 A CN 2013102558766A CN 201310255876 A CN201310255876 A CN 201310255876A CN 103345489 A CN103345489 A CN 103345489A
- Authority
- CN
- China
- Prior art keywords
- auxiliary
- main body
- prescribed information
- event
- information
- 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.)
- Pending
Links
Images
Abstract
The invention discloses an event inquiry demand processing method and device. The event inquiry demand processing method comprises the steps of receiving an event inquiry demand and carrying out word segmentation on a demand text, determining auxiliary limit information of a target event from a word segmentation result using a preset dictionary, wherein the auxiliary limit information comprises time information and/or place information, determining main description information of the target event from the remaining word segmentation results, searching data matched with the main description information and the auxiliary limit information from a preset event data set, and responding the event inquiry demand using the method of searching obtained results. When the method is used for processing event inquiry demands, recall results with high matching degree can be obtained easily, and actual requirements of users can be well met.
Description
Technical field
The present invention relates to technical field of internet application, particularly relate to a kind of event query requests disposal route and device.
Background technology
Along with Internet development, the various functions of network are continually developed, thereby provide convenience for the user in all fields.Be example with the ecommerce, at present, the object of bargain transaction expands to " service " by traditional " commodity ", and the transaction foreground that is characterized in allowing network become and serves under the line is served under this line-transect and just can be solicited customers or business with the line upper type.At present all adopt this mode in a large number as services such as food and drink, tourisms, can effectively save cost for businessman, for the consumer then can utilize network function realize to the service quick screening.
For the website that information on services is provided, the user selects service on line for convenience, and function of search is absolutely necessary.For the query requests of user at " service ", basic processing mode is directly according to the query text of user's input, the content of retrieval and text content coupling in the service commodity information database, this mode is actual identical with common document information retrieval, yet for the inquiry of " service ", but might not be suitable for, for example, the query requests of user's input is " drag in the seabed, Zhong Guan-cun ", purpose is to look for " drag in the seabed " restaurant information that is positioned at " Zhong Guan-cun " area, and do not really want to look for the restaurant of " drag in the seabed, Zhong Guan-cun " by name, if use the inquiry mode of text matches, often be difficult to obtain meeting the result of user's request.
Summary of the invention
For solving the problems of the technologies described above, the embodiment of the invention provides a kind of event query requests disposal route and device, and technical scheme is as follows:
The embodiment of the invention provides a kind of event query requests disposal route, and this method comprises:
Reception event query requests is carried out participle to the request text;
Utilize default dictionary, determine the auxiliary prescribed information of object event from word segmentation result, described auxiliary prescribed information comprises: temporal information and/or location information;
From the residue word segmentation result, determine the main body descriptor of object event;
Concentrate in default event data, retrieve the data of all mating with described main body descriptor and auxiliary prescribed information;
Utilize the retrieval hit results to respond described event query requests.
According to a kind of embodiment of the present invention, described utilization retrieval hit results response events query requests comprises:
According to the correlativity of retrieval hit results with the request text retrieval hit results is sorted, utilize ranking results to respond described event query requests;
Wherein, described correlativity obtains according to the matching degree of main body descriptor and the matching degree weighted calculation of auxiliary prescribed information.
According to a kind of embodiment of the present invention, the weighted value of described weighted calculation determines that by dynamic calculation this Dynamic calculation method comprises:
At the main body descriptor of from the request text, determining and auxiliary prescribed information, obtain described main body descriptor and the auxiliary occurrence number of prescribed information in user's historical query behavior;
According to the occurrence number ratio of main body descriptor and auxiliary prescribed information, determine the weighted value of main body descriptor and auxiliary prescribed information.
According to a kind of embodiment of the present invention, described utilization retrieval hit results response events query requests comprises:
According to main body description field or the auxiliary content that limits field, the retrieval hit results is carried out polymerization, utilize polymerization result to respond described event query requests.
According to a kind of embodiment of the present invention, described according to main body description field or the auxiliary content that limits field, carry out polymerization to retrieving hit results, comprising:
From word segmentation result, only determining under a kind of situation of auxiliary prescribed information, utilizing auxiliary field contents or the main body description field content of limiting of corresponding disappearance, the retrieval hit results is being carried out polymerization.
According to a kind of embodiment of the present invention, this method also comprises:
From word segmentation result, only determining a kind of auxiliary prescribed information, maybe can't determine under the situation of auxiliary prescribed information, utilizing general demand word that the auxiliary prescribed information of disappearance is carried out completion and handle.
The embodiment of the invention also provides a kind of event query requests treating apparatus, it is characterized in that this device comprises:
Word-dividing mode is used for reception event query requests, and the request text is carried out participle;
First information determination module is used for utilizing default dictionary, determines the auxiliary prescribed information of object event from word segmentation result, and described auxiliary prescribed information comprises: temporal information and/or location information;
The second information determination module is used for from the main body descriptor of the definite object event of residue word segmentation result;
Data retrieval module is used for concentrating in default event data, retrieves the data of all mating with described main body descriptor and auxiliary prescribed information;
Respond module is used for utilizing the retrieval hit results to respond described event query requests.
According to a kind of embodiment of the present invention, described respond module specifically is used for:
According to the correlativity of retrieval hit results with the request text retrieval hit results is sorted, utilize ranking results to respond described event query requests;
Wherein, described correlativity obtains according to the matching degree of main body descriptor and the matching degree weighted calculation of auxiliary prescribed information.
According to a kind of embodiment of the present invention, described respond module is determined the weighted value of weighted calculation by dynamic calculation, and this Dynamic calculation method comprises:
At the main body descriptor of from the request text, determining and auxiliary prescribed information, obtain described main body descriptor and the auxiliary occurrence number of prescribed information in user's historical query behavior;
According to the occurrence number ratio of main body descriptor and auxiliary prescribed information, determine the weighted value of main body descriptor and auxiliary prescribed information.
According to a kind of embodiment of the present invention, described respond module specifically is used for:
According to main body description field or the auxiliary content that limits field, the retrieval hit results is carried out polymerization, utilize polymerization result to respond described event query requests.
According to a kind of embodiment of the present invention, described respond module specifically is used for:
From word segmentation result, only determining under a kind of situation of auxiliary prescribed information, utilizing auxiliary field contents or the main body description field content of limiting of corresponding disappearance, the retrieval hit results is being carried out polymerization.
According to a kind of embodiment of the present invention, described first information determination module also is used for:
From word segmentation result, only determining a kind of auxiliary prescribed information, maybe can't determine under the situation of auxiliary prescribed information, utilizing general demand word that the auxiliary prescribed information of disappearance is carried out completion and handle.
The technical scheme that the embodiment of the invention provides, query requests at event, at first utilize participle technique to attempt from ask text, extracting the auxiliary prescribed information in expression " time " or " place ", and then determine simple " event " the main body descriptor that is used for describing own.And then in retrieving, can retrieve as condition with auxiliary prescribed information and main body descriptor two aspects respectively.Compare with the original request text, the text size of auxiliary prescribed information and main body descriptor shortens, thereby the easier result that recalls who obtains high matching degree; On the other hand, request text after the fractionation has more specifically attribute, makes that the specific aim of retrieval is also stronger, has not only effectively improved the retrieval accuracy, and based on the independent retrieval of specific dimension such as " time ", " place ", also can satisfy user's actual demand better.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, the accompanying drawing that describes below only is some embodiment that put down in writing among the present invention, for those of ordinary skills, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is a kind of process flow diagram of embodiment of the invention event query requests disposal route;
Fig. 2 is a kind of structural representation of embodiment of the invention event query requests treating apparatus.
Embodiment
Angle analysis from actual demand, the user is in inquiry service, main pay close attention to following three aspect factor: " time ", " place ", " event ", wherein " event " is the main part that the user pays close attention to, " time " and " place " is then for to main part term restriction in addition.Based on this analysis result, the embodiment of the invention is split as " main body descriptor " and " auxiliary prescribed information " with existing complete service descriptor, sets up the database of corresponding construction and stores.After the user initiates query requests, attempt the query requests text is split as " main body descriptor " and " auxiliary prescribed information " two parts equally, then with these two parts be condition in database, carry out " with " conjunctive search, in order to obtain more and more meet the Query Result of user's request.
Based on above-mentioned principle, a kind of event query requests disposal route that the embodiment of the invention provides can comprise the steps:
Reception event query requests is carried out participle to the request text;
Utilize default dictionary, determine the auxiliary prescribed information of object event from word segmentation result, described auxiliary prescribed information comprises: temporal information and/or location information;
From the residue word segmentation result, determine the main body descriptor of object event;
Concentrate in default event data, retrieve the data of all mating with described main body descriptor and auxiliary prescribed information;
Utilize the retrieval hit results to respond described event query requests.
Such scheme at first utilizes participle technique to attempt extracting the auxiliary prescribed information in expression " time " or " place " from ask text, and then determines simple " event " the main body descriptor that is used for describing own.And then in retrieving, can retrieve as condition with auxiliary prescribed information and main body descriptor two aspects respectively.Compare with the original request text, the text size of auxiliary prescribed information and main body descriptor shortens, thereby the easier result that recalls who obtains high matching degree; On the other hand, request text after the fractionation has more specifically attribute, makes that the specific aim of retrieval is also stronger, has not only effectively improved the retrieval accuracy, and based on the independent retrieval of specific dimension such as " time ", " place ", also can satisfy user's actual demand better.
The executive agent of above-mentioned steps can be an event query requests treating apparatus that is positioned at the inquiry service provider.On the one hand, this device can communicate with user side equipment, in order to receive or respond user's query requests; On the other hand, this device can be visited database and the search operaqtion of execution coupling that event data is stored respectively with " auxiliary prescribed information " and " main body descriptor ".Event query requests treating apparatus can be in the identical hardware entities with database in actual applications, also can be in the different hardware entities, and the embodiment of the invention does not need this to limit.
In order to make those skilled in the art understand technical scheme among the present invention better, below in conjunction with the accompanying drawing in the embodiment of the invention, technical scheme in the embodiment of the invention is described in detail, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, the every other embodiment that those of ordinary skills obtain should belong to the scope that the application protects.
Shown in Figure 1, be the process flow diagram of a kind of event query requests of the present invention disposal route, this method can comprise:
S101 receives the event query requests, and the request text is carried out participle;
When the user need inquire about certain service, the demand of oneself can be described and the input inquiry interface with text mode.In embodiments of the present invention, the query requests text of importing for the user at first carries out word segmentation processing, and the query requests text is divided into several participle unit.
In embodiments of the present invention, the purpose of participle is the fragment that user's request text is split into targetedly a plurality of dissimilar attributes.Be understandable that, here said " participle " must not be defined as certain concrete participle technique, those skilled in the art can select to divide arbitrarily word algorithm according to demand, even when directly in request, importing contents such as punctuate, space as the user, can directly carry out participle according to these natural separators.In order to satisfy the actual needs of inquiry, can also the result of participle be optimized, for example remove stop words etc.Therefore in embodiments of the present invention, do not need " participle " related specific implementation means are limited.
S102 utilizes default dictionary, determines the auxiliary prescribed information of object event from word segmentation result;
The user is the query requests to or a series of events at the query requests of service in essence, in embodiments of the present invention, the event of user's needs inquiry is called " object event ".
For the event description of a macroscopic view, can further be subdivided into several aspects such as " event time ", " location of incident ", " event content " again, can be expressed as " when ", " where " and " what " respectively with English.Wherein " what " is the main part that the user pays close attention to, the main body descriptor that is called event in embodiments of the present invention, temporal information when and location information where then are used for main part term restriction in addition, in embodiments of the present invention temporal information when and location information where are called the auxiliary prescribed information of event.
Generally speaking, the query requests text that the user imports must comprise the main body descriptor, and auxiliary prescribed information then is optional.Wherein, two kinds of auxiliary prescribed informations can be used simultaneously, for example:
Sanya tourism in July, the when:7 month; Where: Sanya; What: tourism.
Two kinds of auxiliary prescribed informations also can be used respectively, for example:
The seabed, Zhong Guan-cun is dragged for and is purchased by group, the when:(sky); Where: Zhong Guan-cun; What: the seabed is dragged for and is purchased by group
June KFC's reward voucher, the when:6 month; The where:(sky); What: KFC's reward voucher
In embodiments of the present invention, query requests text for user's input, at first attempt to determine auxiliary prescribed information wherein, i.e. time information when/ location information where, concrete mode is whether utilize default dictionary to identify each participle unit be temporal information and/or location information.For example, common temporal information form comprises: x, the x month, x day/x number, all x/ week x, and some special addresses such as National Day, Dragon Boat Festival etc.; Common location information comprises administrative division, road, doorplate, bus station, commercial circle, point of interest title etc.Be understandable that those skilled in the art can arrange the content that the identification dictionary comprises arbitrarily according to reality, the embodiment of the invention does not need this to limit.
S103, the main body descriptor of definite object event from the residue word segmentation result;
Here " residue word segmentation result " refers to the part that is not confirmed as auxiliary prescribed information in S102.At S102 by each participle unit of word segmentation result and entry or the template in the dictionary are mated, at first identify the participle unit of wherein representing temporal information and location information, do not have identified residue participle unit, then be identified as the main body descriptor of object event in this step.
S104 concentrates in default event data, retrieves the data of all mating with described main body descriptor and auxiliary prescribed information;
By the processing of S102 and S103, ideal results is to identify one＜when (q) where (q) what (q) from query requests text query〉the tlv triple data.And at the database that is used for storage event data collection, event data also is according to＜when where what〉form store, therefore, respectively with when (q), the where (q) of query requests text correspondence and what (q) as search condition, corresponding field in database carry out " with " conjunctive search, if the when where what field of certain bar data and when (q) where (q) what (q) all can mate, then these data will be called back as hit results.
Be understandable that, here " coupling " should simply not be interpreted as the in full accord of content of text, sensu lato coupling refers to: utilize matching degree that certain algorithm calculates greater than certain preset threshold value, the embodiment of the present application does not need concrete matching algorithm is limited certainly.In addition, actual mate and the process of result screening in, may relate to some special processings, for example synonym/near synonym conversion, semantic conversion, error correction etc. automatically, these all are text retrieval field technological means commonly used, do not need in the embodiment of the present application to be elaborated.
Certainly, in actual application, S102 possibly can't determine auxiliary prescribed information from word segmentation result, perhaps only can determine a kind of auxiliary prescribed information (time or place), the previous case, search condition only comprises what, and the query requests text that in fact is equivalent to directly to utilize the user to import carries out the retrieval based on event content; Latter event, search condition are＜where what〉or＜when what 〉, be equivalent to only use two kinds of conditions to retrieve.
In one embodiment of the invention, for only determining a kind of auxiliary prescribed information or can't determining the situation of auxiliary prescribed information, can utilize general demand word that the auxiliary prescribed information of disappearance is carried out completion and handle.
For example, the query requests text of user's input is " Sanya tourism ", and the result of parsing is:
The when:(sky); Where: Sanya; What: tourism
As seen, the user does not limit the concrete time, but digs certificate by the demand to the user in advance, may obtain some common demands.For example the current time is May, by digging user historical behavior data, can find to have a large number of users to pay close attention to the Sanya travelling products in " June " and " July ", therefore can utilize " June ", " July " query requests text to the user to carry out completion.Again for example, excavate the result according to user's request, the product many places that can find to buy the Sanya tourism then can utilize these concrete time periods to the completion of carrying out of query requests text in some concrete time period.Certainly, actual completion strategy has a variety of, does not need in an embodiment to introduce in detail one by one.In addition, in actual applications, the information of completion can be showed to the user in the mode of recommending, and treats that the user confirms that afterwards the information of recycling completion is retrieved.
S105 utilizes the retrieval hit results to respond described event query requests.
For the result that retrieval is hit, can be left intact and directly show the user.For further applying user's actual demand, also can after doing further processing, hit results show the user again to retrieving.
In a kind of embodiment of the present invention, can sort to the retrieval hit results according to the correlativity of retrieval hit results with the request text, since the pass of a plurality of search conditions be " with " logic, so correlativity can obtain according to the matching degree of main body descriptor and the matching degree weighted calculation of auxiliary prescribed information.
Weighted value can be that ordering system is default, for example according to: when(10%), where(30%), ratio what(60%) is weighted calculating to the degree of correlation of each field.
Weighted value also can determine that concrete grammar is as follows by dynamic calculation:
In advance user's historical query request is added up, for every query requests, statistics is corresponding what and the occurrence number of when/where wherein.Concrete statistical time range can be a nearest week, nearest one month, etc.
For the result that current retrieval is hit, when (q), where (q), the occurrence number ratio of what (q) in user's historical query behavior, the weights of the when where what that will adopt when determining current the ordering.
The meaning of doing like this is: by user's historical behavior, a period of time is more to pay close attention to time, or place, or concrete event content in the past to excavate the user.In fact, excavate the result and show that the main body that the user pays close attention to remains event content, only as auxiliary qualifications, the concern ratio is less relatively for when and where.Therefore in actual applications, above-mentioned default fixedly weights also can come to determine by this method.
In addition, in actual applications, can also further optimize the statistics of " number of times ", for example according to user's historical behavior record, at first " number of clicks/represent number of times " or " buying number of times/number of clicks " of all events added up, choose the forward data of rank as candidate data.From candidate data, extract the occurrence number of corresponding what and when/where then, calculate the foundation of weighted value with statistics as subsequent dynamic.
Be understandable that concrete weighted value is dynamically adjusted strategy can multiple variation, can't describe one by one in embodiments of the present invention that this part content also should not be construed as the restriction to the present invention program.
Except the retrieval hit results is sorted, in another embodiment of the invention, can also carry out polymerization to the retrieval hit results and show.Wherein, the either field among when, where, the what can be as the polymerization field, for example, user search " cuisines purchase by group ", then accordingly result with the following methods polymerization show:
According to the time dimension polymerization:
January: drag in the seabed, little fertile sheep
February: red capital fish, South Beauty
March: South Beauty, drag in the seabed
……
(only be used for schematically illustrating herein, omitted location information)
According to place dimension polymerization:
The Zhong Guan-cun: drag in the seabed, little fertile sheep
Know the spring road: red capital fish, South Beauty
……
(only be used for schematically illustrating herein, omitted temporal information)
According to the polymerization of event dimension:
Drag in the seabed:＜January, Da Zhongsi 〉,＜January, peony garden 〉
Little fertile sheep: in＜February, know the spring road 〉,＜January, Dongzhimen 〉
……
When practical application, the concrete polymerization field that adopts can be that system default is specified, and also can be specified by hand by the user.In a kind of embodiment of the present invention, if from word segmentation result, only determine a kind of auxiliary prescribed information at S102, then can utilize auxiliary field contents or the main body description field content of limiting of corresponding disappearance, the retrieval hit results is carried out polymerization.
For example, the user imports " July, cuisines purchased by group ", and wherein confirmable auxiliary prescribed information is temporal information " July ", the location information disappearance, illustrate that the user is also indeterminate for the requirement in " place ", therefore can show the result to the user intuitively with " where " as the polymerization dimension.In addition, if the event main body descriptor of user's input is fuzzyyer, for example " cuisines ", " film " etc., then Dui Ying result also can be with " what " as the polymerization dimension.
Below introduced respectively scheme is showed in ordering displaying and the polymerization of result for retrieval, when practical application, two kinds of schemes can be used respectively, also can be used in combination, and the embodiment of the invention does not need this to limit.
Corresponding to top method embodiment, the present invention also provides a kind of event query requests treating apparatus, referring to shown in Figure 2, this device can comprise: word-dividing mode 210, first information determination module 220, the second information determination modules 230, data retrieval module 240, respond module 250 describes detailed implementation to each module below:
Word-dividing mode 210 is used for reception event query requests, and the request text is carried out participle;
When the user need inquire about certain service, the demand of oneself can be described and the input inquiry interface with text mode.In embodiments of the present invention, the query requests text of importing for the user at first carries out word segmentation processing, and the query requests text is divided into several participle unit.
In embodiments of the present invention, the purpose of participle is the fragment that user's request text is split into targetedly a plurality of dissimilar attributes.Be understandable that, here said " participle " must not be defined as certain concrete participle technique, those skilled in the art can select to divide arbitrarily word algorithm according to demand, even when directly in request, importing contents such as punctuate, space as the user, can directly carry out participle according to these natural separators.In order to satisfy the actual needs of inquiry, can also the result of participle be optimized, for example remove stop words etc.Therefore in embodiments of the present invention, do not need " participle " related specific implementation means are limited.
First information determination module 220 is used for utilizing default dictionary, determines the auxiliary prescribed information of object event from word segmentation result, and described auxiliary prescribed information comprises: temporal information and/or location information;
The user is the query requests to or a series of events at the query requests of service in essence, in embodiments of the present invention, the event of user's needs inquiry is called " object event ".
For the event description of a macroscopic view, can further be subdivided into several aspects such as " event time ", " location of incident ", " event content " again, can be expressed as " when ", " where " and " what " respectively with English.Wherein " what " is the main part that the user pays close attention to, the main body descriptor that is called event in embodiments of the present invention, temporal information when and location information where then are used for main part term restriction in addition, in embodiments of the present invention temporal information when and location information where are called the auxiliary prescribed information of event.
In embodiments of the present invention, query requests text for user's input, at first attempt to determine auxiliary prescribed information wherein, i.e. time information when/ location information where, concrete mode is whether utilize default dictionary to identify each participle unit be temporal information and/or location information.For example, common temporal information form comprises: x, the x month, x day/x number, all x/ week x, and some special addresses such as National Day, Dragon Boat Festival etc.; Common location information comprises administrative division, road, doorplate, bus station, commercial circle, point of interest title etc.Be understandable that those skilled in the art can arrange the content that the identification dictionary comprises arbitrarily according to reality, the embodiment of the invention does not need this to limit.
The second information determination module 230 is used for from the main body descriptor of the definite object event of residue word segmentation result;
Here " residue word segmentation result " refers to the part that is not confirmed as auxiliary prescribed information in S102.At first information determination module 220, by the entry in word segmentation result and the dictionary or template are mated, at first identify the participle unit of wherein representing temporal information and location information, do not have identified residue participle unit, then be identified as the main body descriptor of object event at the second information determination module 230.
By the processing of first information determination module 220 and the second information determination module 230, ideal results is to identify one＜when (q) where (q) what (q) from query requests text query〉the tlv triple data.And at the database that is used for storage event data collection, event data also is according to＜when where what〉form store, therefore, respectively with when (q), the where (q) of query requests text correspondence and what (q) as search condition, corresponding field in database carry out " with " conjunctive search, if the when where what field of certain bar data and when (q) where (q) what (q) all can mate, then these data will be called back as hit results.
Be understandable that, here " coupling " should simply not be interpreted as the in full accord of content of text, sensu lato coupling refers to: utilize matching degree that certain algorithm calculates greater than certain preset threshold value, the embodiment of the present application does not need concrete matching algorithm is limited certainly.In addition, actual mate and the process of result screening in, may relate to some special processings, for example synonym/near synonym conversion, semantic conversion, error correction etc. automatically, these all are text retrieval field technological means commonly used, do not need in the embodiment of the present application to be elaborated.
Certainly, in actual application, first information determination module 220 possibly can't be determined auxiliary prescribed information from word segmentation result, perhaps only can determine a kind of auxiliary prescribed information (time or place), the previous case, search condition only comprises what, and the query requests text that in fact is equivalent to directly to utilize the user to import carries out the retrieval based on event content; Latter event, search condition are＜where what〉or＜when what 〉, be equivalent to only use two kinds of conditions to retrieve.
In one embodiment of the invention, for only determining a kind of auxiliary prescribed information or can't determining the situation of auxiliary prescribed information, can utilize general demand word that the auxiliary prescribed information of disappearance is carried out completion and handle.
Respond module 250 is used for utilizing the retrieval hit results to respond described event query requests.
For the result that retrieval is hit, can be left intact and directly show the user.For further applying user's actual demand, also can after doing further processing, hit results show the user again to retrieving.
In a kind of embodiment of the present invention, can sort to the retrieval hit results according to the correlativity of retrieval hit results with the request text, since the pass of a plurality of search conditions be " with " logic, so correlativity can obtain according to the matching degree of main body descriptor and the matching degree weighted calculation of auxiliary prescribed information.
Weighted value can be that ordering system is default, for example according to: when(10%), where(30%), ratio what(60%) is weighted calculating to the degree of correlation of each field.
Weighted value also can determine that concrete grammar is as follows by dynamic calculation:
In advance user's historical query request is added up, for every query requests, statistics is corresponding what and the occurrence number of when/where wherein.Concrete statistical time range can be a nearest week, nearest one month, etc.
For the result that current retrieval is hit, when (q), where (q), the occurrence number ratio of what (q) in user's historical query behavior, the weights of the when where what that will adopt when determining current the ordering.
The meaning of doing like this is: by user's historical behavior, a period of time is more to pay close attention to time, or place, or concrete event content in the past to excavate the user.In fact, excavate the result and show that the main body that the user pays close attention to remains event content, only as auxiliary qualifications, the concern ratio is less relatively for when and where.Therefore in actual applications, above-mentioned default fixedly weights also can come to determine by this method.
In addition, in actual applications, can also further optimize the statistics of " number of times ", for example according to user's historical behavior record, at first " number of clicks/represent number of times " or " buying number of times/number of clicks " of all events added up, choose the forward data of rank as candidate data.From candidate data, extract the occurrence number of corresponding what and when/where then, calculate the foundation of weighted value with statistics as subsequent dynamic.
Be understandable that concrete weighted value is dynamically adjusted strategy can multiple variation, can't describe one by one in embodiments of the present invention that this part content also should not be construed as the restriction to the present invention program.
Except the retrieval hit results is sorted, in another embodiment of the invention, can also carry out polymerization to the retrieval hit results and show.Wherein, the either field among when, where, the what can be as the polymerization field, and when practical application, the concrete polymerization field that adopts can be that system default is specified, and also can be specified by hand by the user.In a kind of embodiment of the present invention, if from word segmentation result, only determine a kind of auxiliary prescribed information at first information determination module 220, then can utilize auxiliary field contents or the main body description field content of limiting of corresponding disappearance, the retrieval hit results is carried out polymerization.
Below introduced respectively scheme is showed in ordering displaying and the polymerization of result for retrieval, when practical application, two kinds of schemes can be used respectively, also can be used in combination, and the embodiment of the invention does not need this to limit.
As seen through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize by the mode that software adds essential general hardware platform.Based on such understanding, the part that technical scheme of the present invention contributes to prior art in essence in other words can embody with the form of software product, this computer software product can be stored in the storage medium, as ROM/RAM, magnetic disc, CD etc., comprise that some instructions are with so that a computer equipment (can be personal computer, server, the perhaps network equipment etc.) carry out the described method of some part of each embodiment of the present invention or embodiment.
Each embodiment in this instructions all adopts the mode of going forward one by one to describe, and identical similar part is mutually referring to getting final product between each embodiment, and each embodiment stresses is difference with other embodiment.Especially, for device embodiment, because it is substantially similar in appearance to method embodiment, so describe fairly simplely, relevant part gets final product referring to the part explanation of method embodiment.Device embodiment described above only is schematic, wherein said unit as the separating component explanation can or can not be physically to separate also, the parts that show as the unit can be or can not be physical locations also, namely can be positioned at a place, perhaps also can be distributed on a plurality of network element.Can select wherein some or all of module to realize the purpose of present embodiment scheme according to the actual needs.Those of ordinary skills namely can understand and implement under the situation of not paying creative work.
The above only is the specific embodiment of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (12)
1. event query requests disposal route is characterized in that this method comprises:
Reception event query requests is carried out participle to the request text;
Utilize default dictionary, determine the auxiliary prescribed information of object event from word segmentation result, described auxiliary prescribed information comprises: temporal information and/or location information;
From the residue word segmentation result, determine the main body descriptor of object event;
Concentrate in default event data, retrieve the data of all mating with described main body descriptor and auxiliary prescribed information;
Utilize the retrieval hit results to respond described event query requests.
2. method according to claim 1 is characterized in that, described utilization retrieval hit results response events query requests comprises:
According to the correlativity of retrieval hit results with the request text retrieval hit results is sorted, utilize ranking results to respond described event query requests;
Wherein, described correlativity obtains according to the matching degree of main body descriptor and the matching degree weighted calculation of auxiliary prescribed information.
3. method according to claim 2 is characterized in that, the weighted value of described weighted calculation determines that by dynamic calculation this Dynamic calculation method comprises:
At the main body descriptor of from the request text, determining and auxiliary prescribed information, obtain described main body descriptor and the auxiliary occurrence number of prescribed information in user's historical query behavior;
According to the occurrence number ratio of main body descriptor and auxiliary prescribed information, determine the weighted value of main body descriptor and auxiliary prescribed information.
4. method according to claim 1 is characterized in that, described utilization retrieval hit results response events query requests comprises:
According to main body description field or the auxiliary content that limits field, the retrieval hit results is carried out polymerization, utilize polymerization result to respond described event query requests.
5. method according to claim 4 is characterized in that, and is described according to main body description field or the auxiliary content that limits field, carries out polymerization to retrieving hit results, comprising:
From word segmentation result, only determining under a kind of situation of auxiliary prescribed information, utilizing auxiliary field contents or the main body description field content of limiting of corresponding disappearance, the retrieval hit results is being carried out polymerization.
6. method according to claim 1 is characterized in that, this method also comprises:
From word segmentation result, only determining a kind of auxiliary prescribed information, maybe can't determine under the situation of auxiliary prescribed information, utilizing general demand word that the auxiliary prescribed information of disappearance is carried out completion and handle.
7. event query requests treating apparatus is characterized in that this device comprises:
Word-dividing mode is used for reception event query requests, and the request text is carried out participle;
First information determination module is used for utilizing default dictionary, determines the auxiliary prescribed information of object event from word segmentation result, and described auxiliary prescribed information comprises: temporal information and/or location information;
The second information determination module is used for from the main body descriptor of the definite object event of residue word segmentation result;
Data retrieval module is used for concentrating in default event data, retrieves the data of all mating with described main body descriptor and auxiliary prescribed information;
Respond module is used for utilizing the retrieval hit results to respond described event query requests.
8. device according to claim 7 is characterized in that, described respond module specifically is used for:
According to the correlativity of retrieval hit results with the request text retrieval hit results is sorted, utilize ranking results to respond described event query requests;
Wherein, described correlativity obtains according to the matching degree of main body descriptor and the matching degree weighted calculation of auxiliary prescribed information.
9. device according to claim 8 is characterized in that, described respond module is determined the weighted value of weighted calculation by dynamic calculation, and this Dynamic calculation method comprises:
At the main body descriptor of from the request text, determining and auxiliary prescribed information, obtain described main body descriptor and the auxiliary occurrence number of prescribed information in user's historical query behavior;
According to the occurrence number ratio of main body descriptor and auxiliary prescribed information, determine the weighted value of main body descriptor and auxiliary prescribed information.
10. device according to claim 7 is characterized in that, described respond module specifically is used for:
According to main body description field or the auxiliary content that limits field, the retrieval hit results is carried out polymerization, utilize polymerization result to respond described event query requests.
11. device according to claim 10 is characterized in that, described respond module specifically is used for:
From word segmentation result, only determining under a kind of situation of auxiliary prescribed information, utilizing auxiliary field contents or the main body description field content of limiting of corresponding disappearance, the retrieval hit results is being carried out polymerization.
12. device according to claim 7 is characterized in that, described first information determination module also is used for:
From word segmentation result, only determining a kind of auxiliary prescribed information, maybe can't determine under the situation of auxiliary prescribed information, utilizing general demand word that the auxiliary prescribed information of disappearance is carried out completion and handle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013102558766A CN103345489A (en) | 2013-06-25 | 2013-06-25 | Event inquiry demand processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013102558766A CN103345489A (en) | 2013-06-25 | 2013-06-25 | Event inquiry demand processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103345489A true CN103345489A (en) | 2013-10-09 |
Family
ID=49280284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013102558766A Pending CN103345489A (en) | 2013-06-25 | 2013-06-25 | Event inquiry demand processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103345489A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823868A (en) * | 2014-02-26 | 2014-05-28 | 中国科学院计算技术研究所 | Event recognition method and event relation extraction method oriented to on-line encyclopedia |
| CN104572935A (en) * | 2014-12-30 | 2015-04-29 | 王玉娇 | Retrieval method and device |
| CN107480292A (en) * | 2017-08-29 | 2017-12-15 | 广东电网有限责任公司中山供电局 | A kind of security incident Training Methodology and system based on fuzzy algorithmic approach |
| CN110633406A (en) * | 2018-06-06 | 2019-12-31 | 北京百度网讯科技有限公司 | Event topic generation method and device, storage medium and terminal equipment |
| CN110826735A (en) * | 2019-10-31 | 2020-02-21 | 上海玖道信息科技股份有限公司 | Electric power SCADA intelligent multidimensional query and maintenance method |
| CN111078988A (en) * | 2019-12-23 | 2020-04-28 | 创意信息技术股份有限公司 | Electric power service information hotspot retrieval method and device and electronic equipment |
| WO2021139183A1 (en) * | 2020-01-08 | 2021-07-15 | 百度在线网络技术（北京）有限公司 | Electronic map searching method and device, apparatus, and medium |
| CN113221538A (en) * | 2021-05-19 | 2021-08-06 | 北京百度网讯科技有限公司 | Event library construction method and device, electronic equipment and computer readable medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101350013A (en) * | 2007-07-18 | 2009-01-21 | 北京灵图软件技术有限公司 | Method and system for searching geographical information |
-
2013
- 2013-06-25 CN CN2013102558766A patent/CN103345489A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101350013A (en) * | 2007-07-18 | 2009-01-21 | 北京灵图软件技术有限公司 | Method and system for searching geographical information |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823868A (en) * | 2014-02-26 | 2014-05-28 | 中国科学院计算技术研究所 | Event recognition method and event relation extraction method oriented to on-line encyclopedia |
| CN104572935A (en) * | 2014-12-30 | 2015-04-29 | 王玉娇 | Retrieval method and device |
| CN107480292A (en) * | 2017-08-29 | 2017-12-15 | 广东电网有限责任公司中山供电局 | A kind of security incident Training Methodology and system based on fuzzy algorithmic approach |
| CN110633406A (en) * | 2018-06-06 | 2019-12-31 | 北京百度网讯科技有限公司 | Event topic generation method and device, storage medium and terminal equipment |
| CN110826735A (en) * | 2019-10-31 | 2020-02-21 | 上海玖道信息科技股份有限公司 | Electric power SCADA intelligent multidimensional query and maintenance method |
| CN110826735B (en) * | 2019-10-31 | 2023-04-25 | 上海玖道信息科技股份有限公司 | Electric SCADA intelligent multidimensional query and overhaul method |
| CN111078988A (en) * | 2019-12-23 | 2020-04-28 | 创意信息技术股份有限公司 | Electric power service information hotspot retrieval method and device and electronic equipment |
| WO2021139183A1 (en) * | 2020-01-08 | 2021-07-15 | 百度在线网络技术（北京）有限公司 | Electronic map searching method and device, apparatus, and medium |
| US11609961B2 (en) | 2020-01-08 | 2023-03-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Search method and apparatus for an electronic map, device and medium |
| CN113221538A (en) * | 2021-05-19 | 2021-08-06 | 北京百度网讯科技有限公司 | Event library construction method and device, electronic equipment and computer readable medium |
| CN113221538B (en) * | 2021-05-19 | 2023-09-19 | 北京百度网讯科技有限公司 | Event library construction method and device, electronic equipment and computer readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103345489A (en) | Event inquiry demand processing method and device | |
| JP6301958B2 (en) | Method and apparatus for configuring search terms, delivering advertisements, and retrieving product information | |
| US8037064B2 (en) | Method and system of selecting landing page for keyword advertisement | |
| US20080270333A1 (en) | System and method for determining semantically related terms using an active learning framework | |
| US9727906B1 (en) | Generating item clusters based on aggregated search history data | |
| CN103020049A (en) | Searching method and searching system | |
| CN105677780A (en) | Scalable user intent mining method and system thereof | |
| US20100318427A1 (en) | Enhancing database management by search, personal search, advertising, and databases analysis efficiently using core-set implementations | |
| CN102279851A (en) | Intelligent navigation method, device and system | |
| KR20120073360A (en) | Method and apparatus for geographic specific search results including a map-based display | |
| CN103440286A (en) | Method and system for providing recommended information on the basis of search results | |
| CN111159341B (en) | Information recommendation method and device based on user investment and financial management preference | |
| CN106557480A (en) | Implementation method and device that inquiry is rewritten | |
| EP3485394B1 (en) | Contextual based image search results | |
| CN110111167A (en) | A kind of method and apparatus of determining recommended | |
| CN109074366A (en) | Gain adjustment component for computer network routed infrastructure | |
| CN103257962A (en) | Method and device for providing information | |
| CN103902549A (en) | Search data sorting method and device and data searching method and device | |
| CN105469291A (en) | User information providing method and device | |
| CN107153697A (en) | Product search method and device in a kind of commodity transaction website | |
| US8589234B1 (en) | Companion ad auctions | |
| WO2014159111A2 (en) | Clustering of ads with organic map content | |
| US20090248655A1 (en) | Method and Apparatus for Providing Sponsored Search Ads for an Esoteric Web Search Query | |
| CN101551796A (en) | Control system and corresponding control method for releasing information according to carrier content | |
| CN107085573B (en) | Hotspot information acquisition method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131009 |