CN110569431A - public opinion information monitoring method and device, computer equipment and storage medium - Google Patents
public opinion information monitoring method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110569431A CN110569431A CN201910748248.9A CN201910748248A CN110569431A CN 110569431 A CN110569431 A CN 110569431A CN 201910748248 A CN201910748248 A CN 201910748248A CN 110569431 A CN110569431 A CN 110569431A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- monitoring result
- primary
- user
- result
- 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
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
-
- 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
- G06F16/957—Browsing optimisation, e.g. caching or content distillation
-
- 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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
Abstract
The invention relates to a public opinion information monitoring method, a device, computer equipment and a storage medium, wherein the method comprises the steps of acquiring a monitoring task; carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result; feeding back the primary monitoring result; acquiring the operation behavior of the user on the primary monitoring result to obtain a user behavior log; calculating semantic distances according to the user behavior logs and the primary monitoring result to obtain semantic distances corresponding to all documents in the primary monitoring result; sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result; and sending the target monitoring result to the terminal so that the target monitoring result is displayed on the terminal. According to the public opinion information monitoring method and device, under the condition that a user inputs few monitoring task keyword information, the corresponding monitoring result can be accurately returned, the user experience of public opinion information monitoring is improved, and the public opinion information monitoring accuracy is improved.
Description
Technical Field
the invention relates to a computer, in particular to a public opinion information monitoring method, a public opinion information monitoring device, computer equipment and a storage medium.
background
in recent years, various emergencies occur at some time, and great harm is caused to related enterprises. Because the internet has a huge information scale and rich information but no order, people have to spend a lot of time browsing and searching the information needed by themselves. Public opinion monitoring is an important channel for crisis management of enterprises, and effective network public opinion insight monitoring becomes an important means for the enterprises to resist crisis and reduce crisis. By monitoring the network media public opinion dynamic state, the system can effectively monitor hot events, emergencies, sensitive information and the like on the Internet in real time, and improve the crisis management and control capability.
The public opinion monitoring technology mainly searches network media documents and returns monitoring results based on keyword information by receiving monitoring task keyword information from a user. The traditional public opinion monitoring method is a simple document which literally contains key word information input by a user, and cannot meet the increasingly rich public opinion monitoring requirements of the user, but has the following problems: the monitoring keywords are not treated differently, for example, "apple" can be used for monitoring fruit information and detecting information of an apple computer; monitoring result drift problem, after the monitoring task keywords are expanded, the monitoring result deviates from the original monitoring intention of the user, and the expansion words select irrelevant words; the monitoring keyword information input by the user is limited, and the information requirement of the user cannot be accurately described.
Therefore, it is necessary to design a new method for accurately returning a corresponding monitoring result under the condition that a user inputs less monitoring task keyword information, so as to improve the user experience of public opinion information monitoring and improve the monitoring accuracy of public opinion information.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a public opinion information monitoring method, a public opinion information monitoring device, computer equipment and a storage medium.
in order to achieve the purpose, the invention adopts the following technical scheme: public opinion information monitoring method includes:
acquiring a monitoring task;
Carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
Feeding back the primary monitoring result;
Acquiring the operation behavior of the user on the primary monitoring result to obtain a user behavior log;
Calculating semantic distances according to the user behavior logs and the primary monitoring result to obtain semantic distances corresponding to all documents in the primary monitoring result;
Sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
And sending the target monitoring result to the terminal so that the target monitoring result is displayed on the terminal.
the further technical scheme is as follows: adopt keyword monitoring mode to carry out public opinion monitoring to the monitoring task to obtain first monitoring result, include:
Acquiring a keyword of a monitoring task;
Performing semantic expansion on the keywords of the monitoring task to obtain target keywords;
And searching the network media document based on the target keyword to obtain a primary monitoring result.
the further technical scheme is as follows: the calculating the semantic distance according to the user behavior log and the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result includes:
Analyzing the user behavior log to obtain the user monitoring intention characteristics;
constructing a user preference value according to the user monitoring intention characteristics;
and calculating semantic distances according to the user preference numerical values and the primary monitoring results to obtain the semantic distances corresponding to all documents in the primary monitoring results.
The further technical scheme is as follows: the analyzing the user behavior log to obtain the user monitoring intention characteristics comprises:
Implicit feedback information and explicit feedback information in a user behavior log are extracted;
and expanding the implicit feedback information and the explicit feedback information to obtain the user monitoring intention characteristics.
the further technical scheme is as follows: the constructing of the user preference value according to the user monitoring intention characteristics comprises the following steps:
establishing a corresponding historical interest unit according to the monitoring intention characteristics of the user;
Summing the weights of the historical interest units to obtain a long-term interest value;
Determining a short-term interest value according to the monitoring intention characteristics of the user;
And calculating a user preference value according to the short-term interest value and the long-term interest value.
the further technical scheme is as follows: the calculating the semantic distance according to the user preference value and the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result includes:
determining a corresponding monitoring intention value by using the user preference value and the primary monitoring result;
And calculating the corresponding monitoring intention value and the corresponding primary monitoring result to calculate the correlation so as to obtain the semantic distance corresponding to each document in the primary monitoring result.
The further technical scheme is as follows: the method for sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain the target monitoring result comprises the following steps:
Sequencing each document in the primary monitoring result according to a semantic distance descending arrangement mode corresponding to each document in the primary monitoring result to obtain an intermediate monitoring result;
and screening the documents meeting the requirements from the intermediate monitoring results to form target monitoring results.
the invention also provides a public opinion information monitoring device, comprising:
The task acquisition unit is used for acquiring a monitoring task;
the primary monitoring unit is used for carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
The primary feedback unit is used for feeding back the primary monitoring result;
A behavior log obtaining unit, configured to obtain an operation behavior of the user on the primary monitoring result to obtain a user behavior log;
the distance calculation unit is used for calculating semantic distances according to the user behavior logs and the primary monitoring result so as to obtain semantic distances corresponding to all documents in the primary monitoring result;
the sequencing unit is used for sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
And the sending unit is used for sending the target monitoring result to the terminal so as to display the target monitoring result on the terminal.
The invention also provides computer equipment which comprises a memory and a processor, wherein the memory is stored with a computer program, and the processor realizes the method when executing the computer program.
The invention also provides a storage medium storing a computer program which, when executed by a processor, is operable to carry out the method as described above.
Compared with the prior art, the invention has the beneficial effects that: according to the invention, after primary monitoring is carried out by combining semantic expansion of the monitoring task keywords, a primary monitoring result is output, a user behavior log formed by operation behaviors of the primary monitoring result is analyzed by combining a user, long-term interest and short-term interest of the user are combined with the monitoring task, not only is the correlation between the monitoring task and a document in the monitoring result considered, but also the correlation between the document in the monitoring result and the user interest is considered, rich monitoring requirements of the user can be accurately expressed, the corresponding monitoring result can be accurately returned under the condition that the user inputs less monitoring task keyword information, the user experience of public opinion information monitoring is improved, and the monitoring accuracy of public opinion information is improved.
The invention is further described below with reference to the accompanying drawings and specific embodiments.
drawings
in order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic view of an application scenario of a public opinion information monitoring method according to an embodiment of the present invention;
Fig. 2 is a schematic flow chart illustrating a public opinion information monitoring method according to an embodiment of the present invention;
fig. 3 is a schematic view of a sub-process of a public opinion information monitoring method according to an embodiment of the present invention;
fig. 4 is a schematic view of a sub-process of a public opinion information monitoring method according to an embodiment of the present invention;
fig. 5 is a schematic view of a sub-process of a public opinion information monitoring method according to an embodiment of the present invention;
Fig. 6 is a schematic view of a sub-process of a public opinion information monitoring method according to an embodiment of the present invention;
Fig. 7 is a schematic view of a sub-process of a public opinion information monitoring method according to an embodiment of the present invention;
fig. 8 is a schematic view of a sub-process of a public opinion information monitoring method according to an embodiment of the present invention;
fig. 9 is a schematic block diagram of a public opinion information monitoring device according to an embodiment of the present invention;
FIG. 10 is a schematic block diagram of a computer device provided by an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
it will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
referring to fig. 1 and 2, fig. 1 is a schematic view of an application scenario of a public opinion information monitoring method according to an embodiment of the present invention. Fig. 2 is a schematic flow chart of a public opinion information monitoring method according to an embodiment of the present invention. The public opinion information monitoring method is applied to a server. The server performs data interaction with the terminal, a user sends a monitoring task through the terminal, the server acquires a user behavior log from the terminal after feeding back a primary monitoring result, and a document with higher accuracy is fed back to the terminal again according to the user behavior log and the primary monitoring result. The terminal interacts with the server, the monitoring result is fed back in an interest mode, the quality of the monitoring result can be continuously improved, specifically, a user conducts behavior operation on the primary monitoring result and gives feedback through behavior analysis, the user behavior log is the accumulation of multiple feedback results of the user, and the analysis of the user behavior log is equivalent to the use of a large amount of user-related feedback.
Fig. 2 is a schematic flow chart of a public opinion information monitoring method according to an embodiment of the present invention. As shown in fig. 2, the method includes the following steps S110 to S170.
and S110, acquiring a monitoring task.
in the present embodiment, the monitoring task refers to a word or a sentence or the like for retrieving a document.
and S120, carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result.
In this embodiment, the primary monitoring result refers to a document set formed by performing a web document search using a keyword.
In an embodiment, referring to fig. 3, the step S120 may include steps S121 to S123.
And S121, acquiring keywords of the monitoring task.
In this embodiment, the monitoring task may be analyzed by means of semantic analysis or the like to obtain a keyword of the monitoring task.
And S122, performing semantic expansion on the keywords of the monitoring task to obtain target keywords.
In this embodiment, the target keyword is a word that is consistent with or close to the semantics of the keyword of the monitoring task.
Specifically, the semantic dictionary is used for expanding the keyword semantics, semantic reasoning is introduced in the monitoring process, and the semantic consistency of the user monitoring task and the document is improved.
And S123, retrieving the network media document based on the target keyword to obtain a primary monitoring result.
The keyword expansion of the monitoring task can well distinguish ambiguity through feedback information of a user, and the monitoring effect is improved.
And S130, feeding back the primary monitoring result.
In this embodiment, the initial monitoring result is fed back to the terminal to obtain the operation behavior of the user on the initial monitoring result.
and S140, acquiring the operation behavior of the user on the primary monitoring result to obtain a user behavior log.
In this embodiment, the user behavior log refers to a log of operation behaviors of the document by the user, and the user behavior log may indicate a reflection of the initial monitoring result by the user.
The user behavior log records the user behavior, can truly reflect the preference of the user, analyzes the interest and the preference of the user by the user behavior log, and can more accurately screen out the document meeting the requirements by combining the interest and the preference of the user.
S150, calculating semantic distances according to the user behavior logs and the primary monitoring result to obtain the semantic distances corresponding to all documents in the primary monitoring result.
In this embodiment, the semantic distance refers to a parameter for measuring the relevancy between the monitoring task and the document, where the document is a result obtained by monitoring the monitoring task.
in an embodiment, referring to fig. 4, the step S150 may include steps S151 to S153.
and S151, analyzing the user behavior log to obtain the user monitoring intention characteristics.
in the embodiment, the user monitoring intention characteristic is an intermediate form between the keyword and the real information requirement of the user, and is used for representing the characteristic of the monitoring requirement of the user.
in an embodiment, referring to fig. 5, the step S151 may include steps S1511 to S1512.
S1511, extracting implicit feedback information and explicit feedback information in the user behavior log;
s1512, expanding the implicit feedback information and the explicit feedback information to obtain the user monitoring intention characteristics.
The user monitoring intention characteristics are mainly analyzed through behaviors of a user when the user browses a primary monitoring result page, so that a measure of interest and preference of the user on the primary monitoring result page content is obtained, and user monitoring task information is expanded by utilizing related feedback of an analysis result.
The interest preference of the user to the content of the primary monitoring result is generally divided into two types, one type is implicit feedback interest, namely implicit feedback information in a user behavior log, such as the dwell time of a certain monitoring result page, the click times of a mouse and the length of the page content, and the high click times and the long dwell time of the mouse possibly mean that the user is more interested; the other type is explicit feedback interest, namely explicit feedback information in a user behavior log, and mainly actively marks behaviors of the user, such as collection, approval and forwarding. The long retention time of the page of the primary monitoring result may be caused by the actions of leaving and the like of the user, but if the user has a reading and viewing action, a mouse clicking action is necessarily generated; in addition, the length of the page content of the primary monitoring result also has an important influence on the retention time and the click frequency, so that the ratio of the retention time to the length of the page content of the primary monitoring result can be used for judgment according to actual conditions, and the accuracy of the whole monitoring is further improved.
and S152, constructing a user preference value according to the user monitoring intention characteristics.
in this embodiment, the user preference value refers to a preference degree of the user for each document in the initial monitoring result, and the monitoring intention of the user can be estimated more accurately.
The accurate estimation of the monitoring intention of the user is a key for monitoring the personalized public opinion information, and considering that the monitoring intention is difficult to express clearly enough in the monitoring task submitted by the user, and the operation behavior of the user on the initial monitoring result contains enough user interest information, which is beneficial to accurately understanding the current monitoring intention, therefore, the monitoring task and the user interest need to be considered when modeling the monitoring intention of the user, wherein the user interest comprises short-term interest and long-term interest, the short-term interest represents the transient interest information of the user, the long-term interest represents the continuous interest information, the interest information of the user is accurately described through the combination of the short-term interest and the long-term interest, and the quantification of the user interest can be represented by a user preference value.
in an embodiment, referring to fig. 6, the step S152 may include steps S1521 to S1524.
s1521, establishing a corresponding historical interest unit according to the user monitoring intention characteristics.
In this embodiment, the history interest unit refers to the interest of the user corresponding to different operation behaviors of the primary monitoring result.
S1522, summing the weights of the historical interest units to obtain a long-term interest numerical value.
in this embodiment, the long-term interest value refers to quantized values corresponding to the praise, collection, forwarding and implicit feedback information in the user behavior, that is, quantized values corresponding to the explicit feedback information and the implicit feedback information.
establishing historical interest units for historical monitoring behaviors, wherein each historical interest unit has a corresponding weight, calculating cosine correlation with a monitoring task as the weight of the historical interest unit according to a document set operated by user behaviors of each primary monitoring result, and summing long-term interest values of users by each historical interest unit according to the weight, wherein implicit interest refers to implicit feedback information, namely p (omega | theta)long)＝w1*flike points(d)+ w2*fcollection method(d)+w3*fForwarding(d)+w4*fImplicit interest(d) (ii) a Wherein f isLike points(d)＝log(1+ numNumber of praise)；fCollection method(d)＝log(1+numcollection number)；fForwarding(d)＝log(1+numNumber of forwarding)；ldis the length of the document content, ε is the time-document length threshold coefficient, TC is the total historical visit,Is a historical average access time, w1，w2，w3，w4Weight parameters for offline trainingAnd (4) counting.
S1523, determining a short-term interest value according to the user monitoring intention characteristics.
in this embodiment, the short-term interest value refers to a value formed by a user operating a behavior corresponding to the initial monitoring result, that is, a quantized value of the implicit feedback information. In particular, the amount of the solvent to be used, wherein sgn is a step function, ldIs the length of the document content, t represents the user access dwell time, ε is the time-document length threshold coefficient, p (ω | θ)short) Is a short-term interest value.
S1524, calculating a user preference value according to the short-term interest value and the long-term interest value.
specifically, p (ω | θ) is usedhistory)＝μ*p(ω|θshort)+(1-μ)*p(ω|θlong) Calculating a user preference value, wherein θshortRepresenting interest based on short-term behavior, i.e. short-term interest, p (ω | θ)short) Representing the weight of the word omega in the short-term interest value, thetalongrepresenting interest based on long-term behavior, i.e. long-term interest, p (ω | θ)long) Representing the weight of the word ω in the long-term interest value.
s153, calculating semantic distances according to the user preference values and the primary monitoring result to obtain the semantic distances corresponding to all the documents in the primary monitoring result.
in an embodiment, referring to fig. 7, the step S153 may include steps S1531 to S1532.
S1531, determining a corresponding monitoring intention value by using the user preference value and the primary monitoring result;
In the present embodiment, the monitoring intention value refers to a quantitative value of the intention of the user for the whole monitoring process.
specifically, p' (ω | θ) is adoptedq)＝α*p(ω|θq)+(1-α)*p(ω|θhistory) Calculating a monitoring intention value, wherein ω represents a word and θqkey words, theta, representing user monitoring taskshistoryThe user preference value is shown, and the monitoring intention value p' (ω | θ) can be obtained as described aboveq)＝α*p(ω|θq)+β* p(ω|θshort)+γ*p(ω|θlong) (ii) a Wherein α, β, γ respectively represent weights of the monitoring task keyword, the short-term interest value, and the long-term interest value of the user, and α + β + γ is 1.
in order to determine the weight values of a monitoring task keyword, a short-term interest value and a long-term interest value of a user in a monitoring intention model, linear search is carried out between intervals of 0 to 1 in a 0.1-bit step length, parameters alpha, beta, gamma belong to {0.1,0.2,. eta., 0.9}, and meet alpha + beta + gamma ═ 1, various possible monitoring intention value parameter pairs are tried based on grid search, then cross validation is carried out, namely, a data set is divided into s groups, the s-1 group is randomly extracted as a training set each time, the rest group is used as a test set, n times are randomly extracted to obtain n monitoring intention values, and a weight value combination which enables the MAP (average accuracy rate, Meaneverage prediction) value of the monitoring result of the training set to be maximum is found and set as the value of the parameters alpha, beta and gamma. Wherein the content of the first and second substances, Wherein i is a certain document in the target monitoring result, r is the total number of the target monitoring result documents, and position (i) is the position of the certain document in the target monitoring result, so as to continuously adjust the target monitoring result.
S1532, calculating the corresponding monitoring intention value and the corresponding correlation degree of the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result.
in particular, under the language model framework, semantic distance is often used to measure the relevance of monitoring tasks and documents,Where ω represents a word, V represents the entire vocabulary, q represents the current monitoring task, and θqrepresenting the monitoring intention corresponding to the current monitoring task, d representing a document, thetadRepresenting a document quantization value corresponding to the document d; d (theta)q||θd) Expressing the relevance between the monitoring tasks and the documents, namely semantic distance, and according to the KL distance model, the public opinion monitoring problem is actually equivalent to the monitoring intention theta of respectively estimating the corresponding monitoring tasksqand a document thetadand calculating a monitoring intention theta corresponding to the monitoring taskqAnd a document thetadthe semantic distance between them as the degree of correlation. Under the condition that a user inputs less monitoring task keyword information, the corresponding monitoring result can be accurately returned, the user experience of public opinion information monitoring is improved, and the monitoring accuracy of the public opinion information is improved.
And S160, sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result.
In this embodiment, the target monitoring result refers to a document that meets the user monitoring intention and the monitoring task.
In an embodiment, referring to fig. 8, the step S160 may include steps S161 to S162.
S161, sequencing each document in the primary monitoring result according to a semantic distance descending arrangement mode corresponding to each document in the primary monitoring result to obtain an intermediate monitoring result;
And S162, screening the document meeting the requirements from the intermediate monitoring result to form a target monitoring result.
only under the condition that the relevancy meets the requirement, the target monitoring result can be obtained, a large number of documents which do not meet the requirement are screened, and the user experience feeling of the whole public opinion information monitoring can be improved.
And S170, sending the target monitoring result to a terminal so that the target monitoring result is displayed on the terminal.
the method is realized, the public opinion experimental data come from data sources such as microblogs, WeChat public numbers, blogs, forums, news and the like, the total number of the data sources is about 500, historical data and incremental data are captured by a self-researched distributed crawler tool, the total data amount is about 3500 ten thousand, and the average accuracy MAP index is used as the evaluation index.
the user behavior log can be obtained according to the ID number, date and time of the known user, and the evaluation data statistical information is shown in the following table 1:
and taking the log operation with the behavior of the monitoring result as a related document and the MAP as a monitoring task model judgment standard to measure the accuracy of the monitoring result as a whole. In the experiment, only the model of the original monitoring task is considered to be Mt, the monitoring task model combined with the short-term interest is taken as Mt + short, the monitoring task model combined with the long-term interest is taken as Mt + long, the monitoring task model combined with the long-term interest and the short-term interest is taken as Mt + ensemble, and the public sentiment data is compared with the change of the monitoring performance before and after the combination of the long-term interest and the short-term interest. The results are shown in table 2:
| Model | Mt | Mt+short | Mt+long | Mt+ensemble |
| MAP | 0.3202 | 0.3816 | 0.3903 | 0.4358 |
and only the MAP value of the monitoring result of the monitoring task information is considered to be 0.3202, the monitoring task model of the long-term interest and the short-term interest of the user is fused, and the quality of the monitoring result is improved. Compared with the information monitoring result of the original monitoring task, the MAP is respectively improved by 7.02 percent and 6.14 percent; the monitoring result of the long-time interest and the short-time interest is improved by 11.56%, so that the interest of the user is beneficial to accurately understanding the monitoring intention of the user.
According to the public opinion information monitoring method, after primary monitoring is carried out by combining semantic expansion of monitoring task keywords, a primary monitoring result is output, a user behavior log formed by operation behaviors of the primary monitoring result is analyzed by combining a user, long-term interest and short-term interest of the user are combined with the monitoring task, the correlation between the monitoring task and a document in the monitoring result is considered, the correlation between the document in the monitoring result and the user interest is also considered, rich monitoring requirements of the user can be accurately expressed, the corresponding monitoring result can be accurately returned under the condition that the user inputs less monitoring task keyword information, the user experience of public opinion information monitoring is improved, and the monitoring accuracy of public opinion information is improved.
fig. 9 is a schematic block diagram of a public opinion information monitoring device 300 according to an embodiment of the present invention. As shown in fig. 9, the present invention further provides a public opinion information monitoring device 300 corresponding to the above public opinion information monitoring method. The public opinion information monitoring device 300 includes a unit for performing the above public opinion information monitoring method, and the device may be configured in a server.
specifically, referring to fig. 9, the public opinion information monitoring device 300 includes:
A task obtaining unit 301, configured to obtain a monitoring task;
The primary monitoring unit 302 is used for carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
A primary feedback unit 303, configured to feed back the primary monitoring result;
A behavior log obtaining unit 304, configured to obtain an operation behavior of the user on the primary monitoring result to obtain a user behavior log;
a distance calculating unit 305, configured to calculate semantic distances according to the user behavior log and the primary monitoring result, so as to obtain semantic distances corresponding to each document in the primary monitoring result;
the sorting unit 306 is configured to sort the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
a sending unit 307, configured to send the target monitoring result to the terminal, so that the target monitoring result is displayed on the terminal.
In one embodiment, the primary monitoring unit 302 includes:
the keyword acquisition subunit is used for acquiring keywords of the monitoring task;
The semantic expansion subunit is used for performing semantic expansion on the keywords of the monitoring task to obtain target keywords;
and the retrieval subunit is used for retrieving the network media document based on the target keyword so as to obtain a primary monitoring result.
In one embodiment, the distance calculation unit 305 includes:
The log analysis subunit is used for analyzing the user behavior log to obtain the user monitoring intention characteristics;
The preference construction subunit is used for constructing a user preference value according to the user monitoring intention characteristics;
and the calculating subunit is used for calculating the semantic distance according to the user preference value and the primary monitoring result so as to obtain the semantic distance corresponding to each document in the primary monitoring result.
in one embodiment, the log analysis subunit includes:
The information extraction module is used for extracting implicit feedback information and explicit feedback information in the user behavior log;
And the information extension module is used for extending the implicit feedback information and the explicit feedback information to obtain the monitoring intention characteristics of the user.
In an embodiment, the preference building subunit comprises:
the interest unit establishing module is used for establishing a corresponding historical interest unit according to the monitoring intention characteristics of the user;
The summation module is used for carrying out weight summation on the historical interest units to obtain a long-term interest value;
the short-term interest value determining module is used for determining a short-term interest value according to the monitoring intention characteristics of the user;
And the preference value calculation module is used for calculating the user preference value according to the short-term interest value and the long-term interest value.
In one embodiment, the calculation subunit includes:
The intention value calculation module is used for determining a corresponding monitoring intention value by utilizing a user preference value and the primary monitoring result;
and the relevancy calculation module is used for calculating the corresponding monitoring intention value and the corresponding primary monitoring result to calculate the relevancy so as to obtain the semantic distance corresponding to each document in the primary monitoring result.
in one embodiment, the sorting unit 306 includes:
the descending order arrangement subunit is used for sequencing each document in the primary monitoring result according to a semantic distance descending order arrangement mode corresponding to each document in the primary monitoring result so as to obtain an intermediate monitoring result;
And the screening subunit is used for screening the document meeting the requirement from the intermediate monitoring result to form a target monitoring result.
it should be noted that, as can be clearly understood by those skilled in the art, the detailed implementation process of the public opinion information monitoring device 300 and each unit may refer to the corresponding description in the foregoing method embodiments, and for convenience and brevity of description, no further description is provided herein.
The public opinion information monitoring apparatus 300 may be implemented in a form of a computer program, and the computer program may be executed on a computer device as shown in fig. 10.
Referring to fig. 10, fig. 10 is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 500 may be a server, which may be an independent server or a server cluster composed of a plurality of servers.
referring to fig. 10, the computer device 500 includes a processor 502, memory, and a network interface 505 connected by a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504.
The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 comprises program instructions that, when executed, cause the processor 502 to perform a method for monitoring public opinion information.
the processor 502 is used to provide computing and control capabilities to support the operation of the overall computer device 500.
The internal memory 504 provides an environment for the computer program 5032 in the non-volatile storage medium 503 to run, and when the computer program 5032 is executed by the processor 502, the processor 502 may be enabled to perform a public opinion information monitoring method.
the network interface 505 is used for network communication with other devices. Those skilled in the art will appreciate that the configuration shown in fig. 10 is a block diagram of only a portion of the configuration relevant to the present teachings and is not intended to limit the computing device 500 to which the present teachings may be applied, and that a particular computing device 500 may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
Wherein the processor 502 is configured to run the computer program 5032 stored in the memory to implement the following steps:
acquiring a monitoring task;
carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
feeding back the primary monitoring result;
acquiring the operation behavior of the user on the primary monitoring result to obtain a user behavior log;
calculating semantic distances according to the user behavior logs and the primary monitoring result to obtain semantic distances corresponding to all documents in the primary monitoring result;
Sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
And sending the target monitoring result to the terminal so that the target monitoring result is displayed on the terminal.
In an embodiment, when implementing the step of performing public opinion monitoring on the monitoring task in a keyword monitoring manner to obtain a primary monitoring result, the processor 502 specifically implements the following steps:
acquiring a keyword of a monitoring task;
Performing semantic expansion on the keywords of the monitoring task to obtain target keywords;
and searching the network media document based on the target keyword to obtain a primary monitoring result.
in an embodiment, when the processor 502 implements the step of calculating the semantic distance according to the user behavior log and the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result, the following steps are specifically implemented:
Analyzing the user behavior log to obtain the user monitoring intention characteristics;
constructing a user preference value according to the user monitoring intention characteristics;
And calculating semantic distances according to the user preference numerical values and the primary monitoring results to obtain the semantic distances corresponding to all documents in the primary monitoring results.
In an embodiment, when the processor 502 implements the step of analyzing the user behavior log to obtain the user monitoring intention characteristic, the following steps are specifically implemented:
implicit feedback information and explicit feedback information in a user behavior log are extracted;
And expanding the implicit feedback information and the explicit feedback information to obtain the user monitoring intention characteristics.
In an embodiment, when the processor 502 implements the step of constructing the user preference value according to the user monitoring intention characteristics, the following steps are specifically implemented:
establishing a corresponding historical interest unit according to the monitoring intention characteristics of the user;
Summing the weights of the historical interest units to obtain a long-term interest value;
Determining a short-term interest value according to the monitoring intention characteristics of the user;
and calculating a user preference value according to the short-term interest value and the long-term interest value.
in an embodiment, when the processor 502 implements the step of calculating the semantic distance according to the user preference value and the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result, the following steps are specifically implemented:
Determining a corresponding monitoring intention value by using the user preference value and the primary monitoring result;
And calculating the corresponding monitoring intention value and the corresponding primary monitoring result to calculate the correlation so as to obtain the semantic distance corresponding to each document in the primary monitoring result.
in an embodiment, when the processor 502 implements the step of sorting the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain the target monitoring result, the following steps are specifically implemented:
Sequencing each document in the primary monitoring result according to a semantic distance descending arrangement mode corresponding to each document in the primary monitoring result to obtain an intermediate monitoring result;
and screening the documents meeting the requirements from the intermediate monitoring results to form target monitoring results.
it should be understood that, in the embodiment of the present Application, the Processor 502 may be a Central Processing Unit (CPU), and the Processor 502 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, and the like. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
it will be understood by those skilled in the art that all or part of the flow of the method implementing the above embodiments may be implemented by a computer program instructing associated hardware. The computer program includes program instructions, and the computer program may be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the flow steps of the embodiments of the method described above.
Accordingly, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program, when executed by a processor, causes the processor to perform the steps of:
acquiring a monitoring task;
Carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
Feeding back the primary monitoring result;
Acquiring the operation behavior of the user on the primary monitoring result to obtain a user behavior log;
Calculating semantic distances according to the user behavior logs and the primary monitoring result to obtain semantic distances corresponding to all documents in the primary monitoring result;
Sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
and sending the target monitoring result to the terminal so that the target monitoring result is displayed on the terminal.
In an embodiment, when the processor executes the computer program to implement the step of performing public opinion monitoring on the monitoring task in a keyword monitoring manner to obtain an initial monitoring result, the following steps are specifically implemented:
Acquiring a keyword of a monitoring task;
Performing semantic expansion on the keywords of the monitoring task to obtain target keywords;
And searching the network media document based on the target keyword to obtain a primary monitoring result.
in an embodiment, when the processor executes the computer program to implement the step of calculating the semantic distance according to the user behavior log and the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result, the following steps are specifically implemented:
analyzing the user behavior log to obtain the user monitoring intention characteristics;
constructing a user preference value according to the user monitoring intention characteristics;
and calculating semantic distances according to the user preference numerical values and the primary monitoring results to obtain the semantic distances corresponding to all documents in the primary monitoring results.
in an embodiment, when the processor executes the computer program to implement the step of analyzing the user behavior log to obtain the user monitoring intention characteristic, the following steps are specifically implemented:
implicit feedback information and explicit feedback information in a user behavior log are extracted;
and expanding the implicit feedback information and the explicit feedback information to obtain the user monitoring intention characteristics.
In an embodiment, when the step of constructing the user preference value according to the user monitoring intention characteristics is implemented by the processor executing the computer program, the following steps are specifically implemented:
establishing a corresponding historical interest unit according to the monitoring intention characteristics of the user;
Summing the weights of the historical interest units to obtain a long-term interest value;
determining a short-term interest value according to the monitoring intention characteristics of the user;
and calculating a user preference value according to the short-term interest value and the long-term interest value.
In an embodiment, when the processor executes the computer program to implement the step of calculating the semantic distance according to the user preference value and the primary monitoring result to obtain the semantic distance corresponding to each document in the primary monitoring result, the following steps are specifically implemented:
Determining a corresponding monitoring intention value by using the user preference value and the primary monitoring result;
and calculating the corresponding monitoring intention value and the corresponding primary monitoring result to calculate the correlation so as to obtain the semantic distance corresponding to each document in the primary monitoring result.
In an embodiment, when the processor executes the computer program to implement the step of sorting the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain the target monitoring result, the following steps are specifically implemented:
sequencing each document in the primary monitoring result according to a semantic distance descending arrangement mode corresponding to each document in the primary monitoring result to obtain an intermediate monitoring result;
And screening the documents meeting the requirements from the intermediate monitoring results to form target monitoring results.
the storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a magnetic disk, or an optical disk, which can store various computer readable storage media.
those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
in the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative. For example, the division of each unit is only one logic function division, and there may be another division manner in actual implementation. For example, various elements or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented.
the steps in the method of the embodiment of the invention can be sequentially adjusted, combined and deleted according to actual needs. The units in the device of the embodiment of the invention can be merged, divided and deleted according to actual needs. In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a storage medium. Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a terminal, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. public opinion information monitoring method, its characterized in that includes:
Acquiring a monitoring task;
carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
feeding back the primary monitoring result;
acquiring the operation behavior of the user on the primary monitoring result to obtain a user behavior log;
Calculating semantic distances according to the user behavior logs and the primary monitoring result to obtain semantic distances corresponding to all documents in the primary monitoring result;
sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
And sending the target monitoring result to the terminal so that the target monitoring result is displayed on the terminal.
2. the public opinion information monitoring method according to claim 1, wherein the public opinion monitoring is performed by adopting a keyword monitoring mode for a monitoring task to obtain a primary monitoring result, and the method comprises the following steps:
acquiring a keyword of a monitoring task;
Performing semantic expansion on the keywords of the monitoring task to obtain target keywords;
and searching the network media document based on the target keyword to obtain a primary monitoring result.
3. The public opinion information monitoring method according to claim 1, wherein the calculating semantic distances according to the user behavior logs and the primary monitoring results to obtain semantic distances corresponding to each document in the primary monitoring results comprises:
Analyzing the user behavior log to obtain the user monitoring intention characteristics;
Constructing a user preference value according to the user monitoring intention characteristics;
And calculating semantic distances according to the user preference numerical values and the primary monitoring results to obtain the semantic distances corresponding to all documents in the primary monitoring results.
4. the public opinion information monitoring method according to claim 3, wherein the analyzing the user behavior log to obtain the user monitoring intention characteristics comprises:
Implicit feedback information and explicit feedback information in a user behavior log are extracted;
And expanding the implicit feedback information and the explicit feedback information to obtain the user monitoring intention characteristics.
5. the public opinion information monitoring method according to claim 4, wherein the constructing of the user preference value according to the user monitoring intention characteristics comprises:
Establishing a corresponding historical interest unit according to the monitoring intention characteristics of the user;
summing the weights of the historical interest units to obtain a long-term interest value;
determining a short-term interest value according to the monitoring intention characteristics of the user;
And calculating a user preference value according to the short-term interest value and the long-term interest value.
6. The public opinion information monitoring method according to claim 5, wherein the calculating semantic distances according to the user preference values and the primary monitoring results to obtain semantic distances corresponding to each document in the primary monitoring results comprises:
determining a corresponding monitoring intention value by using the user preference value and the primary monitoring result;
and calculating the corresponding monitoring intention value and the corresponding primary monitoring result to calculate the correlation so as to obtain the semantic distance corresponding to each document in the primary monitoring result.
7. the public opinion information monitoring method according to claim 1, wherein the ranking of the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain the target monitoring result comprises:
Sequencing each document in the primary monitoring result according to a semantic distance descending arrangement mode corresponding to each document in the primary monitoring result to obtain an intermediate monitoring result;
and screening the documents meeting the requirements from the intermediate monitoring results to form target monitoring results.
8. Public opinion information monitoring devices, its characterized in that includes:
The task acquisition unit is used for acquiring a monitoring task;
the primary monitoring unit is used for carrying out public opinion monitoring on the monitoring task in a keyword monitoring mode to obtain a primary monitoring result;
The primary feedback unit is used for feeding back the primary monitoring result;
A behavior log obtaining unit, configured to obtain an operation behavior of the user on the primary monitoring result to obtain a user behavior log;
The distance calculation unit is used for calculating semantic distances according to the user behavior logs and the primary monitoring result so as to obtain semantic distances corresponding to all documents in the primary monitoring result;
the sequencing unit is used for sequencing the documents in the primary monitoring result according to the semantic distance corresponding to each document in the primary monitoring result to obtain a target monitoring result;
And the sending unit is used for sending the target monitoring result to the terminal so as to display the target monitoring result on the terminal.
9. a computer device, characterized in that the computer device comprises a memory, on which a computer program is stored, and a processor, which when executing the computer program implements the method according to any of claims 1 to 7.
10. a storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910748248.9A CN110569431A (en) | 2019-08-14 | 2019-08-14 | public opinion information monitoring method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910748248.9A CN110569431A (en) | 2019-08-14 | 2019-08-14 | public opinion information monitoring method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110569431A true CN110569431A (en) | 2019-12-13 |
Family
ID=68775171
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910748248.9A Pending CN110569431A (en) | 2019-08-14 | 2019-08-14 | public opinion information monitoring method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110569431A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112818234A (en) * | 2021-02-02 | 2021-05-18 | 中慧绿浪科技(天津)集团有限公司 | Network public opinion information analysis processing method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007002727A2 (en) * | 2005-06-28 | 2007-01-04 | Claria Corporation | Method for providing advertising content to an internet user based on the user's demonstrated content preferences |
| CN103235777A (en) * | 2012-12-28 | 2013-08-07 | 国家计算机网络与信息安全管理中心 | Method for obtaining user demand-oriented internet public opinion |
| CN103593425A (en) * | 2013-11-08 | 2014-02-19 | 南方电网科学研究院有限责任公司 | Preference-based intelligent retrieval method and system |
| CN107133290A (en) * | 2017-04-19 | 2017-09-05 | 中国人民解放军国防科学技术大学 | A kind of Personalized search and device |
-
2019
- 2019-08-14 CN CN201910748248.9A patent/CN110569431A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007002727A2 (en) * | 2005-06-28 | 2007-01-04 | Claria Corporation | Method for providing advertising content to an internet user based on the user's demonstrated content preferences |
| CN103235777A (en) * | 2012-12-28 | 2013-08-07 | 国家计算机网络与信息安全管理中心 | Method for obtaining user demand-oriented internet public opinion |
| CN103593425A (en) * | 2013-11-08 | 2014-02-19 | 南方电网科学研究院有限责任公司 | Preference-based intelligent retrieval method and system |
| CN107133290A (en) * | 2017-04-19 | 2017-09-05 | 中国人民解放军国防科学技术大学 | A kind of Personalized search and device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112818234A (en) * | 2021-02-02 | 2021-05-18 | 中慧绿浪科技(天津)集团有限公司 | Network public opinion information analysis processing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cherfi et al. | Very fast C4. 5 decision tree algorithm | |
| Walker | Sampling the Dirichlet mixture model with slices | |
| CN107862022B (en) | Culture resource recommendation system | |
| US9727605B1 (en) | Query language identification | |
| CN109033408B (en) | Information pushing method and device, computer readable storage medium and electronic equipment | |
| US20110313844A1 (en) | Real-time-ready behavioral targeting in a large-scale advertisement system | |
| US20190146984A1 (en) | Prioritizing survey text responses | |
| Sharma et al. | NIRMAL: Automatic identification of software relevant tweets leveraging language model | |
| CN107153656B (en) | Information searching method and device | |
| CN104933100A (en) | Keyword recommendation method and device | |
| CN109582862B (en) | Click rate estimation method, medium, system and computing device | |
| JP6678666B2 (en) | Offline evaluation of ranking functions | |
| Zhang et al. | Click-based evidence for decaying weight distributions in search effectiveness metrics | |
| US9996529B2 (en) | Method and system for generating dynamic themes for social data | |
| CN111723260A (en) | Method and device for acquiring recommended content, electronic equipment and readable storage medium | |
| Sosunov et al. | The impact of hospital size on CMS hospital profiling | |
| Lv et al. | An exploration of ranking heuristics in mobile local search | |
| JP5952756B2 (en) | Prediction server, program and method for predicting future number of comments in prediction target content | |
| US20140129694A1 (en) | Evaluating information retrieval systems in real-time across dynamic clusters of evidence | |
| CN110569431A (en) | public opinion information monitoring method and device, computer equipment and storage medium | |
| US8745042B2 (en) | Determining matching degrees between information categories and displayed information | |
| Jia et al. | Handling big data of online social networks on a small machine | |
| CN111414455A (en) | Public opinion analysis method, device, electronic equipment and readable storage medium | |
| CN111382331A (en) | Method, device and system for processing public sentiment topics based on big data | |
| Changxin | Cultivation of college students' employment and entrepreneurship ability of agriculture and forestry engineering major under the background of internet plus |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191213 |