CN110110235B - Method and device for pushing information - Google Patents
Method and device for pushing information Download PDFInfo
- Publication number
- CN110110235B CN110110235B CN201910401497.0A CN201910401497A CN110110235B CN 110110235 B CN110110235 B CN 110110235B CN 201910401497 A CN201910401497 A CN 201910401497A CN 110110235 B CN110110235 B CN 110110235B
- Authority
- CN
- China
- Prior art keywords
- sequence
- keywords
- relevance
- probability
- fusion
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Abstract
The embodiment of the disclosure discloses a method and a device for processing data, and a specific implementation mode of the method comprises the following steps: respectively determining a relevancy sequence by utilizing at least two preset models based on a probability graph, wherein the relevancy sequence is used for representing relevancy of a pre-generated keyword and objects in a pre-acquired object set; fusing the determined association degree sequence to obtain a fused association degree sequence; and determining the related objects of the keywords from the object set based on the fusion relevance sequence. The method for processing the data can improve the accuracy of determining the associated object of the keyword, avoid unreasonable prediction results caused by only using a model based on a probability map for data processing, and further can use the obtained associated object for information push, so that a terminal user can conveniently obtain data processing results, and user experience is improved.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of computers, in particular to a method and a device for pushing information.
Background
The probability graph-based model for data processing is suitable for various technical fields, and the main factors influencing the accuracy of data processing comprise: the algorithm structure of the model based on the probability map and the type of the input parameters are single, and the input parameters are not classified in the related art, so that the accuracy of the result of data processing is low.
Disclosure of Invention
The embodiment of the disclosure provides a method and a device for pushing information.
In a first aspect, an embodiment of the present disclosure provides a method for pushing information, where the method includes: respectively determining a relevancy sequence by utilizing at least two preset models based on a probability graph, wherein the relevancy sequence is used for representing relevancy of a pre-generated keyword and objects in a pre-acquired object set; fusing the determined association degree sequence to obtain a fused association degree sequence; and determining the related objects of the keywords from the object set based on the fusion relevance sequence.
In some embodiments, before determining the sequence of relevance degrees respectively by using at least two preset probability map-based models, the method further includes: classifying the keywords generated in advance to obtain a first category of keywords and a second category of keywords, so that the model based on the probability graph takes the first category of keywords and the second category of keywords as input parameters to determine the association degree sequence.
In some embodiments, before determining the sequence of relevance degrees respectively by using at least two preset probability map-based models, the method further includes: and acquiring the prior probability of the objects in the object set, so that the model based on the probability graph takes the prior probability of the objects as an input parameter to determine the association degree sequence.
In some embodiments, the preset at least two probability map-based models include: two probability map models in series or two probability map models in parallel; and respectively determining the association degree sequence by utilizing at least two preset models based on the probability map, wherein the association degree sequence comprises the following steps: inputting a first type of keywords and prior probabilities into a first of two probability map models connected in series for calculation, and inputting an output result of the first probability map model and second type of keywords into a second of the two probability map models connected in series for calculation to obtain a first association degree sequence; inputting one of the first type keyword and the second type keyword and prior probability into one of two probability map models connected in parallel for calculation, inputting the other of the first type keyword and the second type keyword and prior probability into the other of the two probability map models connected in parallel for calculation, summing output results of the two probability map models, and normalizing the summed result to obtain a second relevancy sequence.
In some embodiments, fusing the determined sequence of relevance to obtain a fused sequence of relevance, includes: and determining an average value or a weighted average value of the relevance degrees corresponding to the same object in the first relevance degree sequence and the second relevance degree sequence, and taking the determined average value or the weighted average value as the fusion relevance degree to obtain a fusion relevance degree sequence.
In some embodiments, fusing the determined sequence of relevance to obtain a fused sequence of relevance, includes: and counting the number of the relevance degrees indicating any object in the first relevance degree sequence and the second relevance degree sequence, and taking the number as the fusion relevance degree to obtain a fusion relevance degree sequence.
In some embodiments, determining the associated object of the keyword from the object set based on the fused association degree sequence includes: and determining an object corresponding to the maximum fusion association degree as an associated object of the keyword.
In some embodiments, after determining the associated object of the keyword from the set of objects, the method further comprises: and filtering the determined associated objects of the keywords based on a preset rule.
In a second aspect, an embodiment of the present disclosure provides an apparatus for pushing information, including: the association degree determining unit is configured to respectively determine an association degree sequence by utilizing at least two preset probability map-based models, wherein the association degree sequence is used for representing the association degrees of the pre-generated keywords and the objects in the pre-acquired object set; the fusion unit is configured to fuse the determined association degree sequence to obtain a fusion association degree sequence; and the related object determining unit is configured to determine the related objects of the keywords from the object set based on the fusion relevance sequence.
In some embodiments, the apparatus further comprises: and the keyword classification unit is configured to classify the pre-generated keywords, obtain the first category of keywords and the second category of keywords, and determine the association degree sequence by taking the first category of keywords and the second category of keywords as input parameters based on the model of the probability map.
In some embodiments, the apparatus further comprises: and a prior probability acquiring unit for acquiring the prior probability of the objects in the object set so that the model based on the probability graph takes the prior probability of the objects as an input parameter to determine the association degree sequence.
In some embodiments, the preset at least two probability map-based models include: two probability map models in series or two probability map models in parallel; and the association degree determination unit is further configured to: inputting a first type of keywords and prior probabilities into a first of two probability map models connected in series for calculation, and inputting an output result of the first probability map model and second type of keywords into a second of the two probability map models connected in series for calculation to obtain a first association degree sequence; inputting one of the first type keyword and the second type keyword and prior probability into one of two probability map models connected in parallel for calculation, inputting the other of the first type keyword and the second type keyword and prior probability into the other of the two probability map models connected in parallel for calculation, summing output results of the two probability map models, and normalizing the summed result to obtain a second relevancy sequence.
In some embodiments, the fusion unit is further configured to: and calculating the average value or weighted average value of the relevance degrees corresponding to the same object in the first relevance sequence and the second relevance sequence, and taking the average value or weighted average value as the fusion relevance degree to obtain a fusion relevance degree sequence.
In some embodiments, the fusion unit is further configured to: and counting the number of the relevance degrees indicating any one object in the first relevance degree sequence and the second relevance degree sequence, and taking the number as fusion relevance degrees to obtain a fusion relevance degree sequence.
In some embodiments, the associated object determination unit is further configured to: and determining an object corresponding to the maximum fusion association degree as an associated object of the keyword.
In some embodiments, the apparatus further comprises: and the filtering unit is configured to filter the determined related objects of the keywords based on a preset rule.
In a third aspect, an embodiment of the present disclosure provides a server, including: one or more processors; a storage device having one or more programs stored thereon; when executed by one or more processors, cause the one or more processors to implement a method as in the first aspect.
In a fourth aspect, embodiments of the present disclosure provide a computer readable medium having stored thereon a computer program which, when executed by a processor, implements the method of the first aspect.
According to the method and the device for pushing the information, firstly, the association degree sequences are respectively determined by utilizing at least two preset probability map-based models, multiple association degree sequence prediction results based on the same original data can be obtained, then the determined association degree sequences are fused to obtain a fused association degree sequence, finally, the associated objects of the keywords are determined from the object set based on the fused association degree sequence, the accuracy of determining the associated objects of the keywords can be improved, unreasonable prediction results caused by data processing only by utilizing one probability map-based model are avoided, further, the obtained associated objects can be utilized for information pushing, so that a terminal user can conveniently obtain data processing results, and user experience is improved.
Drawings
Other features, objects and advantages of the disclosure will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is an exemplary system architecture diagram in which some embodiments of the present disclosure may be applied;
FIG. 2 is a flow diagram for one embodiment of a method for pushing information, according to the present disclosure;
FIG. 3 is a flow diagram of yet another embodiment of a method for pushing information according to the present disclosure;
FIG. 4 is a flow diagram of yet another embodiment of a method for pushing information according to the present disclosure;
FIG. 5 is a schematic diagram of one embodiment of a probability map-based model for a method of pushing information, according to an embodiment of the present disclosure;
FIG. 6 is a schematic diagram of another embodiment of a probability map-based model for a method of pushing information, in accordance with an embodiment of the present disclosure;
FIG. 7 is a flow diagram of one application scenario of a method for pushing information in accordance with an embodiment of the present disclosure;
FIG. 8 is a schematic block diagram illustrating one embodiment of an apparatus for pushing information in accordance with the present disclosure;
FIG. 9 is a schematic structural diagram of an electronic device suitable for use in implementing embodiments of the present disclosure.
Detailed Description
The present disclosure is described in further detail below with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1 illustrates an exemplary system architecture 100 of a method for pushing information or an apparatus for pushing information to which embodiments of the present disclosure may be applied.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user 110 may use the terminal devices 101, 102, 103 to interact with the server 105 over the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have various communication client applications installed thereon, such as a search engine application, a shopping application, an instant messaging tool, a mailbox client, social platform software, a video playing application, and the like.
The terminal apparatuses 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices having a display screen, including but not limited to smart phones, tablet computers, e-book readers, laptop portable computers, desktop computers, and the like. When the terminal apparatuses 101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., multiple pieces of software or software modules to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.
The server 105 may be a server that provides various services. For example, the server 105 may be a background server providing support for the terminal devices 101, 102, 103. The background server may analyze, store, or calculate data submitted by the terminal, and push a data prediction result determined by using the model based on the probability map to the terminal devices 101, 102, and 103.
Generally, the method for pushing information provided by the embodiments of the present disclosure is generally performed by the server 105, and accordingly, the apparatus for pushing information is generally disposed in the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to fig. 2, a flow 200 of one embodiment of a method for pushing information in accordance with the present disclosure is shown. The method for pushing the information comprises the following steps:
in step 210, a sequence of relevance degrees is respectively determined by using at least two preset probability map-based models, wherein the sequence of relevance degrees is used for characterizing the relevance degrees of the pre-generated keywords and the pre-acquired objects in the object set.
In this embodiment, the algorithm of the method 200 for pushing information includes at least two probability map-based models (the executing subject running the algorithm may be the server 105 shown in fig. 1), each probability map-based model takes a pre-generated keyword and a pre-acquired object in an object set as input parameters, and calculates a conditional probability of the object in the object set under the condition of the keyword, where the conditional probability is the association degree.
As an implementation manner of this embodiment, the number of objects in the object set is not limited, that is, one object may be used, or a plurality of objects may be used. Correspondingly, the number of the preset keywords is not limited, that is, one keyword or a plurality of keywords may be used. Based on the above embodiment, the conditional probability may be a joint probability distribution of the objects under the condition that multiple keywords occur simultaneously, and the conditional probabilities calculated by multiple objects form a sequence of relevance degrees.
The probability graph-based model is calculated based on a bayesian probability formula, specifically, the conditional probability of the keyword under the condition of the object and the prior probability of the object are obtained, and then the conditional probability or the joint distribution of the conditional probability of the object under the condition of the keyword is calculated through a series of formula derivation based on the bayesian formula.
In step 220, fusing the determined association degree sequence to obtain a fused association degree sequence;
in this embodiment, the fusion may be, for example, a fusion of the relevance degrees indicating the same object in the relevance degree sequence calculated by different models based on the probability map.
For example, assume that there are two association degree sequences a ═ a1，a2，a3}；B＝{b1，b2，b3In which a1、b2As a corresponding object c1Degree of association of (a)2、b1As a corresponding object c2Degree of association of (a)3、b3As a corresponding object c3The degree of association of (c).
Since the degree of association is a conditional probability, in some alternative embodiments, the fusion may be a probabilistic fusion, i.e., object c1The fusion association degree of (a) may be1+b2) /2, object c2The fusion association degree of (a) may be2+b1) /2, object c1The fusion association degree of (a) may be3+b3) And 2, the fusion association degree sequence is { (a)1+b2)/2，(a2+b1)/2，(a3+b3)/2}。
In other alternative embodiments, the two association degree sequences a ═ { a ═ are used1，a2，a3}；B＝{b1，b2，b3In which a1、b1As a corresponding object c1Degree of association of (a)2As a corresponding object c2Degree of association of (a), (b)2As a corresponding object c3Degree of association of (a)3As a corresponding object c4Degree of association of (a), (b)3As a corresponding object c5The degree of association of (c). The fusion may be to count the number of relevancy degrees corresponding to each object, specifically, the object c1Has a fusion association degree of 2, object c2～c5If the fusion association degrees of (1) are all 1, the fusion association degree sequence is {2, 1, 1, 1, 1 }.
In step 230, based on the fused association degree sequence, an associated object of the keyword is determined from the object set.
In this embodiment, the maximum value of the relevancy in the fusion relevancy sequence may be selected as a screening rule to determine the relevancy object of the keyword.
Referring to the embodiment in step 220, assume that the fused relevance sequence is { (a)1+b2)/2，(a2+b1)/2，(a3+b3) In/2, the degree of fusion association (a)1+b2) The maximum 2 corresponds to the fusion association degree (a)1+b2) Object c of/21It is determined as the associated object of the keyword.
Or in the statistical fusion association degree sequence, the number of the association degrees of any object is indicated, and the object with the largest number is determined as the associated object of the keyword.
With continued reference to the embodiment in step 220, assuming that the fusion association degree sequence is {2, 1, 1, 1, 1}, the object c corresponding to the fusion association degree 2 is1The associated object determined as a keyword.
Furthermore, in some optional implementations, in addition to the above step 210 to step 230, the method for pushing information of the present disclosure may further include: and carrying out information push based on the determined related objects of the keywords.
In these alternative implementations, the execution subject may feed back the associated object of the determined keyword to the terminal device in the form of information push. Further, the execution main body can also push the push information related to the associated object to the terminal equipment according to the determined associated object. Furthermore, the method of the embodiment can determine the associated object associated with the keyword, so that the push information generated based on the associated object has stronger pertinence, and the problem of network resource waste possibly caused by pushing information with low association degree to the terminal is avoided as much as possible.
The following describes a method for pushing information according to an embodiment of the present disclosure, taking vehicle failure diagnosis as an example.
The application scenario of the automobile fault diagnosis is described as follows:
the pre-generated key being a fault in the vehicle, e.g. key k1: flameout during driving; key word k2: the fire is not fired again. The pre-acquired object being the cause of a malfunction in the vehicle, e.g. object ob1: the temperature environment of the water temperature sensing device is abnormal; object ob2: abnormality of gasoline pump; object ob3: a throttle position sensor abnormality; object ob4: the timing belt broke. Determining an object set OB ═ OB by using prior knowledge in the technical field of the existing automobiles1，ob2，ob3，ob4The prior probability of any one of them, i.e. the probability without considering any condition, is obtained to obtain the prior probability set P of the objectr＝{pr1，pr2，pr3，pr4Fourthly, acquiring an inducing outline of the fault reason to the faultRate Pin＝{pin11，pin12，pin13，pin14，pin21，pin22，pin23，pin24In which p isin11＝(k1|ob1)，pin12＝(k1|ob2)，pin13＝(k1|ob3)，pin14＝(k1|ob4)；pin21＝(k2|ob1)，pin22＝(k2|ob2)，pin23＝(k2|ob3)，pin24＝(k2|ob4)。
The execution subject performs step 210, the prior probability PrInduction probability PinRespectively inputting at least two models based on probability graphs for calculation, and obtaining the conditional probability p (ob) of the object in the object set under the condition of keywordsi|kj) Wherein ob isi∈OB，i∈{1，2，3，4}，j∈{1，2}。
Further, the execution principal utilizes a probability map-based model with a conditional probability p (ob)i|kj) Calculating parameters to obtain conditional probability joint distribution p (ob) of objects in the object set under the condition of multiple keywordsi|k1，k2) I.e., the degree of association, the plurality of degrees of association constitute a sequence of degrees of association.
The execution subject executes step 220 and step 230 in turn, and obtains a joint probability distribution p (ob) based on models of different probability mapsi|k1，k2) And fusing to obtain a fusion association degree sequence, and then determining the associated object of the keyword determined from the object set according to the fusion association degree sequence, namely determining the reason causing the automobile fault.
According to the method for pushing the information, firstly, the preset at least two models based on the probability map are used for respectively determining the relevance sequence, and a plurality of relevance sequence prediction results based on the same original data can be obtained. And then, fusing the determined association degree sequence to obtain a fused association degree sequence. And finally, determining the related objects of the keywords from the object set based on the fusion relevance sequence. The accuracy of determining the related objects of the keywords can be improved, and unreasonable prediction results caused by data processing by using only one model based on a probability map are avoided. Furthermore, the obtained associated object can be utilized to carry out information push, so that a terminal user can conveniently obtain a data processing result, and the user experience is improved.
With further reference to fig. 3, a flow 300 of yet another embodiment of a method for pushing information according to the present disclosure is shown. The flow 300 of the method for pushing information includes the following steps:
in step 310, the pre-generated keywords are classified to obtain keywords of a first type and keywords of a second type.
In the present embodiment, the text information input by the user through the terminal device may be preprocessed (e.g., text parsing, recognition, feature extraction, etc.) by a natural language processing unit in an execution subject (e.g., the server 105 shown in fig. 1) to obtain a keyword generated in advance. Furthermore, the natural language processing unit can classify the pre-generated keywords to obtain a first category of keywords and a second category of keywords, and the relevance sequence is determined by taking the first category of keywords and the second category of keywords as input parameters based on the model of the probability map.
In step 320, prior probabilities of objects in the set of objects are obtained.
In this embodiment, the prior probability of the subject represents the probability of the subject without other condition, and taking the subject as a disease as an example, the prior probability of the subject may be the incidence of the disease.
The prior probability of the object is also used as an input parameter of the probability map-based model, i.e. the probability map-based model is calculated by taking at least one of the first category keyword and the second category keyword and the prior probability as the input parameter to determine the associated object of the keyword from the object set.
The prior probabilities of the objects may be stored on an executing agent (e.g., the server 105 in fig. 1) or on other electronic devices communicatively coupled to the executing agent, and when the prior probabilities of the objects are stored on other electronic devices communicatively coupled to the executing agent, the executing agent may obtain the prior probabilities of the corresponding objects by sending data request instructions to the electronic devices.
In step 330, a sequence of relevance degrees is respectively determined by using at least two preset probability map-based models, wherein the sequence of relevance degrees is used for characterizing the relevance degrees of the pre-generated keywords and the pre-acquired objects in the object set.
In step 340, the determined association degree sequences are fused to obtain a fused association degree sequence.
In step 350, based on the fused association degree sequence, an associated object of the keyword is determined from the object set.
The steps 330 to 350 can be performed in a similar manner to the steps 210 to 230 in the embodiment shown in fig. 2, and are not described herein again.
According to the method for pushing information provided by the above embodiment of the disclosure, by classifying the pre-generated keywords and using the classified keywords as the input parameters of the model based on the probability map, the hierarchical calculation of the model based on the probability map can be realized, so that the relevancy sequence is determined based on different types of keywords, and the accuracy of determining the keyword relevancy object is further improved.
With further reference to fig. 4, a flow 400 of yet another embodiment of a method for pushing information according to the present disclosure is shown. The flow 400 of the method for pushing information comprises the following steps:
in step 410, a sequence of relevance degrees is respectively determined by using at least two preset probability map-based models, wherein the sequence of relevance degrees is used for representing the relevance degrees of the pre-generated keywords and the pre-acquired objects in the object set.
In step 420, fusing the determined association degree sequence to obtain a fused association degree sequence;
in step 430, based on the fused association degree sequence, an associated object of the keyword is determined from the object set.
The steps 410 to 430 may be performed in a similar manner to the steps 210 to 230 in the embodiment shown in fig. 2, and are not described herein again.
In step 441, the first type keyword and the prior probability are input into a first one of two probability map models (PGM) connected in series for calculation, and the output result of the first probability map model and the second type keyword are input into a second one of the two probability map models connected in series for calculation, so as to obtain a first relevancy sequence.
In the present embodiment, the probability map-based model running in the execution subject (e.g., the server 105 shown in fig. 1) includes two probability map models connected in series, and as shown in fig. 5, the probability map model shown by reference numeral 510 in the probability map-based model may be referred to as a parent PGM, and the probability map model shown by reference numeral 520 may be referred to as a child PGM.
When the execution subject performs the association degree calculation, as an implementation manner of this embodiment, the probabilistic graph model 510 uses the first category keyword and the prior probability as input parameters to determine the association degree between the first keyword and the object in the object set. The probability map model 520 takes the relevance between the first keyword and the object in the object set and the second keyword determined by the probability map model 510 as input parameters, and determines the relevance between the first keyword and the object in the object set and the relevance between the second keyword and the object in the object set, so as to obtain a first relevance sequence.
This step 441 implements a hierarchical PGM model, and the subsequent child PGM increases the accuracy of determining the object associated with the keyword based on the first association degree sequence with the output result of the parent PGM as the prior probability.
In step 442, one of the first category keyword and the second category keyword and the prior probability are input into one of the two probability map models connected in parallel for calculation, the other of the first category keyword and the second category keyword and the prior probability are input into the other of the two probability map models connected in parallel for calculation, the output results of the two probability map models are summed, and the summed result is normalized to obtain a second correlation sequence.
In the present embodiment, the probability map-based model running in the execution subject (for example, the server 105 shown in fig. 1) includes two probability map models connected in parallel, and as shown in fig. 6, as an implementation manner of the present embodiment, the probability map model shown by reference numeral 610 takes a keyword of a first type and a prior probability as input parameters, and determines a degree of association between the keyword and an object in the object set; the probability graph model 620 shown by reference numeral 620 takes the second category of keywords and the prior probability as input parameters, determines the association degree between the second category of keywords and the objects in the object set, then, the execution subject sums the association degree between the first keyword and the objects in the object set and the association degree between the second category of keywords and the objects in the object set, and normalizes the summation result, thereby obtaining a second association degree sequence.
The step 442 also implements a hierarchical probability map model, and through the association degrees determined by the PGMs based on different types of keywords and the summation and normalization processing of the association degree results determined by the PGMs, the result deviation caused by uncertain factors of a single PGM can be eliminated, and the accuracy of determining the object associated with the keyword can be improved.
It should be noted that, the steps 441 and 442 are only two preferred embodiments of the present disclosure, and do not limit the technical solution of the present disclosure in any way.
In step 451, an average value or a weighted average value of the degrees of association corresponding to the same object in the first degree of association sequence and the second degree of association sequence is calculated, and the average value or the weighted average value is used as the fusion degree of association to obtain a fusion degree of association sequence.
In this embodiment, the method for pushing information, which is executed in an execution subject (for example, the server 105 shown in fig. 1), further includes fusing the association degree sequences determined by the multiple probability map-based models to eliminate errors of different probability map-based models and reduce the influence of uncertainty factors on the accuracy of determining the object associated with the keyword.
As in the foregoing embodiment of the present disclosure, the relevance is a conditional probability of an object in the object set under the condition of the keyword, and therefore, an average value or a weighted average value of the relevance corresponding to the same object in the first relevance sequence and the second relevance sequence is calculated, and in fact, an average value or a weighted average value of the conditional probability corresponding to the same object in the two relevance sequences is calculated.
Specifically, in this embodiment, the number of the models based on the probability map is two, so that the value of the weight is between (0,2), and the accuracy is one bit after the decimal point.
In step 452, the number of relevance degrees indicating any object in the first relevance degree sequence and the second relevance degree sequence is counted, and the number is used as the fusion relevance degree to obtain a fusion relevance degree sequence.
In this embodiment, the number of statistical relevance degrees is used as the fusion relevance degree, which has reliability and feasibility in a statistical sense, and the calculation cost is low.
In step 460, it is determined that the object corresponding to the maximum value of the fused relevance is the relevant object of the keyword.
In this embodiment, as an implementation manner, the maximum value of the average value or the weighted average value of the relevance degrees, that is, the maximum value of the conditional probability corresponding to a certain object, indicates that the object is most likely to be related to the keyword in the object set.
In another embodiment, the number of relevance degrees corresponding to an object is the largest, which statistically indicates that the object is most likely to be relevant to the keyword.
In step 440, the determined associated objects of the keywords are filtered based on the preset rules.
The determined objects are filtered by using the preset rules, and unreasonable results appearing in the objects determined by the model based on the probability map can be further filtered, so that the accuracy of determining the related objects of the keywords in the object set is improved, namely, the accuracy of the prediction result is improved.
It should be noted that the preset rule is preset by the expert in the related art according to the knowledge of the field to which the expert belongs, and the preset rule can be programmed to run in the execution subject (for example, the server 105 shown in fig. 1).
As an alternative embodiment, the method for pushing information of the present disclosure may be applied to auxiliary medical diagnosis, and the preset rule may be preset by a medical expert according to medical knowledge, and is used for filtering strong symptom exclusion diseases, for example, male may not suffer from gynecological diseases, cough may not be acute bronchitis in 20 years, and the like.
According to the method for pushing the information, the model based on the probability map can realize the calculation of the layered probability map model, and the association degree sequences determined by different probability map models are fused, so that the error of the model of the individual probability map can be eliminated, the accuracy of determining the associated object of the keyword is improved, and the determined associated object is filtered based on the preset rule to filter out unreasonable results in the object, so that the accuracy of prediction is further improved.
With continued reference to fig. 7, fig. 7 is a flow 700 of an application scenario of a method for pushing information according to the present embodiment. The process 700 of the application scenario includes the following steps:
The execution subject (e.g., server 105 shown in fig. 1) receives the medical record text sent by the terminal. A user (e.g., a physician) enters medical record text through a terminal (e.g., terminals 101, 102, 103 shown in fig. 1), which can be an electronic medical record.
The execution main body comprises a natural language processing unit, an intelligent algorithm is integrated in the natural language processing unit, the execution main body analyzes the received medical record text by using the natural language processing unit, and the disease is extracted based on the analysis result.
Step 703 to step 704.
The natural language processing unit of the executing subject executes step 703 and step 704, classifies the disorders, obtains positive disorders and negative disorders, i.e., implements the extraction of positive disorders and the extraction of negative disorders. Wherein a positive disorder is a disorder that has already occurred in the patient and a negative disorder is a disorder that has not occurred in the patient.
For example, the medical records are entered as follows: intermittent right epigastric pain for more than ten years, relapse aggravating for two months; with nausea, vomiting; has no fever, chills, hematemesis and dark stool. Wherein the positive disorder comprises: right upper abdominal pain, nausea, vomiting; negative-going disorders include: fever, chills, hematemesis, and dark stool.
Step 705 and step 706.
The natural language processing unit of the execution main body classifies based on positive symptoms and negative symptoms to obtain symptoms (first type keywords) and signs (second type keywords), wherein the symptoms comprise positive symptoms and negative symptoms, the positive symptoms are symptoms appearing in the patient, and the negative symptoms are symptoms not appearing in the patient; the signs include positive signs and negative signs, wherein the positive signs are signs appearing in the patient, and the negative signs are signs not appearing in the patient.
And step 707, acquiring disease incidence.
In the present embodiment, the executing subject sends a data request to a data storage unit (for example, the server 106 shown in fig. 1) based on the symptom (first category keyword) and the sign (second category keyword) to obtain the incidence (prior probability) of the disease that can cause the symptom and the sign; meanwhile, the executive body also respectively obtains a first inducing probability of the positive symptoms under the disease occurrence condition, a first inhibiting probability of the negative symptoms under the disease occurrence condition, a second inducing probability of the positive signs under the disease occurrence condition and a second inhibiting probability of the negative signs under the disease occurrence condition.
Step 708 to step 710.
In this embodiment, three models based on a probability map are run in the execution body, and a unified strategy, a parallel strategy and a series strategy are respectively executed.
The unified strategy comprises a PGM, and the unified strategy has no classification requirement on input parameters, namely, a first induction probability, a first inhibition probability, a second induction probability, a second inhibition probability and a prior probability are directly input into the PGM executing the unified strategy for calculation so as to obtain a relevancy sequence corresponding to the unified strategy;
the parallel strategy comprises two PGMs which are associated, and the parallel strategy has classification requirements on input parameters, namely, a first induction probability, a first inhibition probability and a prior probability are input into one of the two PGMs for calculation, a second induction probability, a second inhibition probability and a prior probability are input into the other one of the two PGMs for calculation, the association degree sequences obtained by calculation of the two PGMs are summed, and the summed result is normalized to obtain the association degree sequence corresponding to the parallel strategy.
The series strategy comprises two PGMs connected in series, and the series strategy has classification requirements on input parameters, namely, a first induction probability, a first inhibition probability and a prior probability are input into a parent PGM for calculation, the output of the parent PGM is used as the prior probability input of a child PGM, and a second induction probability and a second inhibition probability are input into the child PGM for calculation to obtain a correlation sequence corresponding to the series strategy.
And step 711, executing fusion.
The execution main body fuses the association degree sequence corresponding to the unified strategy, the association degree sequence corresponding to the parallel strategy and the association degree sequence corresponding to the series strategy to obtain a fused association degree sequence, wherein the fused association degree sequence can be, for example, joint probability distribution of each disease under the condition that multiple positive symptoms, negative symptoms, positive signs and negative signs occur simultaneously, the steps of the fusion method are the same as those in the foregoing embodiment, and are not repeated here.
And determining the disease corresponding to the maximum value in the fusion probability sequence as a prediction result based on the fusion probability sequence.
Step 712 to step 714.
Further, the execution main body filters the prediction result based on a preset rule to eliminate unreasonable disease reasoning, and feeds the filtered prediction result back to the terminal, namely, informs the user of a diagnosis result based on the current medical record.
The preset rule can be preset by a medical expert according to medical knowledge and stored in the data storage unit, and the execution subject can acquire the preset rule by sending a data request.
As an embodiment, the preset rule may be set based on the gender and duration of symptoms.
It should be understood that the above-described exemplary application scenario of the method for pushing information is shown in fig. 7, and does not represent a limitation of the present disclosure.
With further reference to fig. 8, as an implementation of the methods shown in the above-mentioned figures, the present disclosure provides an embodiment of a method for pushing information, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus may be applied to various electronic devices in particular.
As shown in fig. 8, the apparatus 800 for pushing information may include: a relevancy determining unit 810 configured to determine relevancy sequences by using at least two preset probability map-based models, wherein the relevancy sequences are used for representing relevancy between a pre-generated keyword and an object in a pre-acquired object set; a fusion unit 820 configured to fuse the determined association degree sequence to obtain a fusion association degree sequence; the associated object determining unit 830 is configured to determine an associated object of the keyword from the object set based on the fused association degree sequence.
In some optional implementations of this embodiment, the apparatus 800 for pushing information further includes: the keyword classification unit 840 is configured to classify the pre-generated keywords, obtain the first category keywords and the second category keywords, and determine the association degree sequence by using the first category keywords and the second category keywords as input parameters based on the model of the probability map.
In some optional implementations of this embodiment, the apparatus 800 for pushing information further includes: a prior probability obtaining unit 850 configured to obtain a prior probability of the object in the object set, so that the model based on the probability map determines the association degree sequence with the prior probability of the object as an input parameter.
In some optional implementations of this embodiment, the preset at least two probability map-based models include: two probabilistic graphical models in series or two probabilistic graphical models in parallel, the relevance determining unit 810 is further configured to: inputting a first type of keywords and prior probabilities into a first of two probability map models connected in series for calculation, and inputting an output result of the first probability map model and second type of keywords into a second of the two probability map models connected in series for calculation to obtain a first association degree sequence; inputting one of the first type keyword and the second type keyword and prior probability into one of two probability map models connected in parallel for calculation, inputting the other of the first type keyword and the second type keyword and prior probability into the other of the two probability map models connected in parallel for calculation, summing output results of the two probability map models, and normalizing the summed result to obtain a second relevancy sequence.
In some optional implementations of this embodiment, the fusion unit 820 is further configured to: and calculating the average value or weighted average value of the relevance degrees corresponding to the same object in the first relevance sequence and the second relevance sequence, and taking the average value or weighted average value as the fusion relevance degree to obtain a fusion relevance degree sequence.
In some optional implementations of this embodiment, the fusion unit 820 is further configured to: and counting the number of the relevance degrees indicating any object in the first relevance degree sequence and the second relevance degree sequence, and taking the number as the fusion relevance degree to obtain a fusion relevance degree sequence.
In some optional implementations of this embodiment, the associated object determining unit 830 is further configured to: and determining an object corresponding to the maximum fusion association degree as an associated object of the keyword.
In some optional implementations of this embodiment, the apparatus 800 for pushing information further includes:
and a filtering unit 860 configured to filter the associated objects of the determined keywords based on a preset rule.
Referring now to FIG. 9, shown is a schematic diagram of an electronic device (e.g., the server of FIG. 1) 900 suitable for use in implementing embodiments of the present disclosure. The server shown in fig. 9 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 9, the electronic device 900 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 901 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)902 or a program loaded from a storage means 908 into a Random Access Memory (RAM) 903. In the RAM903, various programs and data necessary for the operation of the electronic apparatus 900 are also stored. The processing apparatus 901, the ROM 902, and the RAM903 are connected to each other through a bus 904. An input/output (I/O) interface 905 is also connected to bus 904.
Generally, the following devices may be connected to the I/O interface 905: input devices 906 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 907 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; and a communication device 909. The communication device 909 may allow the electronic apparatus 900 to perform wireless or wired communication with other apparatuses to exchange data. While fig. 9 illustrates an electronic device 900 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 9 may represent one device or may represent multiple devices as desired.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication device 909, or installed from the storage device 908, or installed from the ROM 902. The computer program, when executed by the processing apparatus 901, performs the above-described functions defined in the methods of the embodiments of the present disclosure. It should be noted that the computer readable medium described in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In embodiments of the present disclosure, however, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied by the server; or may exist separately and not be assembled into the server. The computer readable medium carries one or more programs which, when executed by the server, cause the server to: respectively determining a relevancy sequence by utilizing at least two preset models based on a probability graph, wherein the relevancy sequence is used for representing relevancy of a pre-generated keyword and objects in a pre-acquired object set; fusing the determined association degree sequence to obtain a fused association degree sequence; and determining the related objects of the keywords from the object set based on the fusion relevance sequence.
Computer program code for carrying out operations for embodiments of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware. The described units may also be provided in a processor, and may be described as: a processor includes a degree-of-association determination unit, a fusion unit, and an associated object determination unit. The names of the units do not constitute a limitation to the units themselves in some cases, for example, the association degree determination unit may also be described as "a unit that determines the association degree sequence respectively using at least two probability map-based models set in advance".
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.
Claims (18)
1. A method for pushing information, comprising:
respectively determining a relevancy sequence by utilizing at least two preset models based on a probability graph, wherein the relevancy sequence is used for representing relevancy of a pre-generated keyword and objects in a pre-acquired object set;
fusing the determined association degree sequence to obtain a fused association degree sequence;
determining the related objects of the keywords from the object set based on the fusion relevance sequence;
carrying out information push based on the determined related objects of the keywords;
the fusion of the determined association degree sequence to obtain a fusion association degree sequence comprises the following steps:
and fusing the relevance degrees used for indicating the same object in the relevance degree sequence calculated by different models based on the probability graph.
2. The method according to claim 1, wherein before the determining the sequence of relevance degrees respectively by using at least two preset probability map-based models, the method further comprises:
classifying the keywords generated in advance to obtain a first category of keywords and a second category of keywords, so that the relevance sequence is determined by taking the first category of keywords and the second category of keywords as input parameters based on a model of a probability graph.
3. The method according to claim 2, wherein before the determining the sequence of relevance degrees respectively by using at least two preset probability map-based models, the method further comprises:
and acquiring the prior probability of the objects in the object set, so that the model based on the probability graph takes the prior probability of the objects as an input parameter to determine the association degree sequence.
4. The method of claim 3, wherein the preset at least two probability map-based models comprise: two probability map models in series or two probability map models in parallel; and
the determining the association degree sequence by utilizing at least two preset models based on the probability map respectively comprises the following steps:
inputting the first type of keywords and the prior probability into a first of two probability map models connected in series for calculation, and inputting the output result of the first probability map model and the second type of keywords into a second of the two probability map models connected in series for calculation to obtain a first relevancy sequence;
and calculating one of the first category keywords and the second category keywords and one of the two probability map models with the prior probability input in parallel, calculating the other of the first category keywords and the second category keywords and the other of the two probability map models with the prior probability input in parallel, summing output results of the two probability map models, and normalizing the summed result to obtain a second correlation sequence.
5. The method of claim 4, wherein the fusing the determined sequence of relevance to obtain a fused sequence of relevance comprises:
and determining an average value or a weighted average value of the relevance degrees of the corresponding same object in the first relevance degree sequence and the second relevance degree sequence, and taking the determined average value or the weighted average value as a fusion relevance degree to obtain a fusion relevance degree sequence.
6. The method of claim 4, wherein the fusing the determined sequence of relevance to obtain a fused sequence of relevance comprises:
and counting the number of the relevance degrees indicating any one object in the first relevance degree sequence and the second relevance degree sequence, and taking the number as fusion relevance degrees to obtain a fusion relevance degree sequence.
7. The method according to claim 5 or 6, wherein the determining the associated object of the keyword from the object set based on the fused association degree sequence comprises:
and determining the object corresponding to the maximum value of the fusion association degree as the associated object of the keyword.
8. The method according to any of claims 1-6, wherein after said determining the associated object of the keyword from the set of objects, the method further comprises:
and filtering the determined associated objects of the keywords based on a preset rule.
9. An apparatus for pushing information, comprising:
the association degree determining unit is configured to respectively determine an association degree sequence by utilizing at least two preset probability map-based models, wherein the association degree sequence is used for representing the association degrees of the pre-generated keywords and the objects in the pre-acquired object set;
the fusion unit is configured to fuse the determined association degree sequence to obtain a fusion association degree sequence;
an associated object determining unit configured to determine an associated object of the keyword from the object set based on the fused association degree sequence;
a unit configured to perform information pushing based on the determined related objects of the keywords;
the fusion unit is further configured to:
and fusing the relevance degrees used for indicating the same object in the relevance degree sequence calculated by different models based on the probability graph.
10. The apparatus of claim 9, wherein the apparatus further comprises:
the keyword classification unit is configured to classify the pre-generated keywords, obtain a first category of keywords and a second category of keywords, and determine the relevancy sequence by using the first category of keywords and the second category of keywords as input parameters based on a model of a probability map.
11. The apparatus of claim 10, wherein the apparatus further comprises:
a prior probability obtaining unit configured to obtain a prior probability of an object in the object set, so that the model based on the probability map determines the association degree sequence with the prior probability of the object as an input parameter.
12. The apparatus of claim 11, wherein the preset at least two probability map-based models comprise: two probability map models in series or two probability map models in parallel; and
the association degree determination unit is further configured to:
inputting the first type of keywords and the prior probability into a first of two probability map models connected in series for calculation, and inputting the output result of the first probability map model and the second type of keywords into a second of the two probability map models connected in series for calculation to obtain a first relevancy sequence;
and calculating one of the first category keywords and the second category keywords and one of the two probability map models with the prior probability input in parallel, calculating the other of the first category keywords and the second category keywords and the other of the two probability map models with the prior probability input in parallel, summing output results of the two probability map models, and normalizing the summed result to obtain a second correlation sequence.
13. The apparatus of claim 12, wherein the fusion unit is further configured to:
and calculating the average value or weighted average value of the relevance degrees of the corresponding same object in the first relevance degree sequence and the second relevance degree sequence, and taking the average value or weighted average value as the fusion relevance degree to obtain a fusion relevance degree sequence.
14. The apparatus of claim 12, wherein the fusion unit is further configured to:
and counting the number of the relevance degrees indicating any one object in the first relevance degree sequence and the second relevance degree sequence, and taking the number as fusion relevance degrees to obtain a fusion relevance degree sequence.
15. The apparatus according to claim 13 or 14, wherein the associated object determination unit is further configured to:
and determining the object corresponding to the maximum value of the fusion association degree as the associated object of the keyword.
16. The apparatus of any of claims 9-14, wherein the apparatus further comprises:
and the filtering unit is configured to filter the determined related objects of the keywords based on a preset rule.
17. A server for pushing information, comprising:
one or more processors;
a storage device having one or more programs stored thereon;
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-8.
18. A computer-readable medium, on which a computer program is stored, wherein the program, when executed by a processor, implements the method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401497.0A CN110110235B (en) | 2019-05-14 | 2019-05-14 | Method and device for pushing information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401497.0A CN110110235B (en) | 2019-05-14 | 2019-05-14 | Method and device for pushing information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110110235A CN110110235A (en) | 2019-08-09 |
| CN110110235B true CN110110235B (en) | 2021-08-31 |
Family
ID=67490119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910401497.0A Active CN110110235B (en) | 2019-05-14 | 2019-05-14 | Method and device for pushing information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110110235B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105956130A (en) * | 2016-05-09 | 2016-09-21 | 浙江农林大学 | Multi-information fusion scientific research literature theme discovering and tracking method and system thereof |
| CN106570046A (en) * | 2016-03-02 | 2017-04-19 | 合网络技术（北京）有限公司 | Method and device for recommending relevant search data based on user operation behavior |
| CN106777304A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | The method for pushing and device of theme |
| CN206741482U (en) * | 2017-03-27 | 2017-12-12 | 乐山师范学院 | A kind of network public-opinion supervising device using probability graph model |
| WO2018009643A1 (en) * | 2016-07-07 | 2018-01-11 | Aspen Technology, Inc. | Computer systems and methods for performing root cause analysis and building a predictive model for rare event occurrences in plant-wide operations |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9710730B2 (en) * | 2011-02-11 | 2017-07-18 | Microsoft Technology Licensing, Llc | Image registration |
| GB201209382D0 (en) * | 2012-05-25 | 2012-07-11 | Poikos Ltd | Body measurement |
| KR102275452B1 (en) * | 2017-03-16 | 2021-07-12 | 한국전자통신연구원 | Method for tracking image in real time considering both color and shape at the same time and apparatus therefor |
| CN109656541A (en) * | 2018-11-20 | 2019-04-19 | 东软集团股份有限公司 | Exploitative recommended method, device, storage medium and electronic equipment |
-
2019
- 2019-05-14 CN CN201910401497.0A patent/CN110110235B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570046A (en) * | 2016-03-02 | 2017-04-19 | 合网络技术（北京）有限公司 | Method and device for recommending relevant search data based on user operation behavior |
| CN105956130A (en) * | 2016-05-09 | 2016-09-21 | 浙江农林大学 | Multi-information fusion scientific research literature theme discovering and tracking method and system thereof |
| WO2018009643A1 (en) * | 2016-07-07 | 2018-01-11 | Aspen Technology, Inc. | Computer systems and methods for performing root cause analysis and building a predictive model for rare event occurrences in plant-wide operations |
| CN106777304A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | The method for pushing and device of theme |
| CN206741482U (en) * | 2017-03-27 | 2017-12-12 | 乐山师范学院 | A kind of network public-opinion supervising device using probability graph model |
Non-Patent Citations (1)
| Title |
|---|
| 《异构信息网络数据上的融合概率图模型》;吴蕾等;《http://www.cnki.net/kcms/doi/10.3778/j.issn.1673-9418.1401005.html》;20140303;第712-718页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110110235A (en) | 2019-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11710071B2 (en) | Data analysis and rendering | |
| US11232365B2 (en) | Digital assistant platform | |
| US10395330B2 (en) | Evaluating vendor communications for accuracy and quality | |
| US10817521B2 (en) | Near-real-time prediction, classification, and notification of events in natural language systems | |
| US20180011979A1 (en) | Question generation systems and methods for automating diagnosis | |
| CN114616560A (en) | Techniques for adaptive and context-aware automation service composition for Machine Learning (ML) | |
| US20220035998A1 (en) | Obtaining supported decision trees from text for medical health applications | |
| CN112256886B (en) | Probability calculation method and device in atlas, computer equipment and storage medium | |
| CN117131281B (en) | Public opinion event processing method, apparatus, electronic device and computer readable medium | |
| US11341394B2 (en) | Diagnosis of neural network | |
| CN110110235B (en) | Method and device for pushing information | |
| Bublyk et al. | Decision Support System Design For Low-Voice Emergency Medical Calls At Smart City Based On Chatbot Management In Social Networks. | |
| CN113221762A (en) | Cost balance decision method, insurance claim settlement decision method, device and equipment | |
| CN114219663A (en) | Product recommendation method and device, computer equipment and storage medium | |
| Boytcheva et al. | Integrating Data Analysis Tools for Better Treatment of Diabetic Patients. | |
| US20230412639A1 (en) | Detection of malicious on-chain programs | |
| CN113077353B (en) | Method, device, electronic equipment and medium for generating nuclear insurance conclusion | |
| CN116469576B (en) | Dispute event identification method, device, electronic equipment and computer readable medium | |
| US11973832B2 (en) | Resolving polarity of hosted data streams | |
| US11977515B1 (en) | Real time analysis of interactive content | |
| CN113077352B (en) | Insurance service article recommending method based on user information and insurance related information | |
| US20230237180A1 (en) | Systems and methods for linking a screen capture to a user support session | |
| CN111753080B (en) | Method and device for outputting information | |
| US20230043849A1 (en) | Answer generation using machine reading comprehension and supported decision trees | |
| CN116932269A (en) | Fault processing method, device, electronic equipment and computer storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |