CN109597874B - Information recommendation method, device and server - Google Patents
Information recommendation method, device and server Download PDFInfo
- Publication number
- CN109597874B CN109597874B CN201811238927.3A CN201811238927A CN109597874B CN 109597874 B CN109597874 B CN 109597874B CN 201811238927 A CN201811238927 A CN 201811238927A CN 109597874 B CN109597874 B CN 109597874B
- Authority
- CN
- China
- Prior art keywords
- reference information
- information
- preset
- target reference
- target
- 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
Abstract
The embodiment of the invention discloses an information recommendation method, an information recommendation device and a server, wherein the method comprises the following steps: analyzing the search information input by a user to obtain target reference information matched with the search information, wherein the similarity between the search information and the target reference information is larger than a preset similarity threshold; acquiring the historical call times of the target reference information, wherein the historical call times are times of matching of the target reference information with at least one search information input by a user in a first preset time period; determining recommendation information according to the historical call times; and sending the recommendation information to the client. By implementing the method, the content possibly interested by the user can be analyzed by combining the historical statistical data, so that more accurate portraits are carried out on the user, and the accuracy of the recommendation result is improved.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to an information recommendation method, an information recommendation device, and a server.
Background
The intelligent question-answering is to orderly and scientifically arrange the accumulated unordered corpus problems and establish a classification model based on knowledge; the classification models can guide the newly added corpus consultation and service problems, save human resources, improve the automaticity of problem processing and reduce the running cost of websites.
The related question recommendation is one of the core functions of the intelligent question and answer, when a user puts forward a question to the intelligent question and answer system, the system pushes out not only the question answer, but also knowledge related to the question for the user to inquire, so that all questions are comprehensively mastered by one question. A commonly used related problem recommendation method is to actively recommend problems or services that may be of interest to a user based on the current user's problems and historical user portraits. However, the method cannot accurately portray the user for a new knowledge base or a new user, so that the user is not interested in the related problem of recommendation, and the purpose of recommendation function is not achieved.
Disclosure of Invention
The embodiment of the invention provides an information recommendation method, which can be used for more accurately portraying a user and improving the accuracy of a recommendation result.
In a first aspect, an embodiment of the present invention provides an information recommendation method, where the method includes:
analyzing the search information input by a user to obtain target reference information matched with the search information, wherein the similarity between the search information and the target reference information is larger than a preset similarity threshold;
Acquiring the historical call times of the target reference information, wherein the historical call times are times of matching of the target reference information with at least one search information input by a user in a first preset time period;
determining recommendation information according to the historical call times;
and sending the recommendation information to the client.
In a second aspect, the present invention provides an information recommendation apparatus, the apparatus comprising:
the analysis module is used for analyzing the search information input by the user to obtain target reference information matched with the search information, and the similarity between the search information and the target reference information is larger than a preset similarity threshold;
the acquisition module is used for acquiring the historical call times of the target reference information, wherein the historical call times are times of matching of the target reference information and at least one search information input by a user in a first preset time period;
the determining module is used for determining recommendation information according to the historical call times;
and the sending module is used for sending the recommendation information to the client.
In the embodiment of the invention, a user inputs search information in a client, the client sends the search information to a server, the server analyzes the search information input by the user to obtain target reference information matched with the search information, the server detects through a core word to find recommendation information similar to the target reference information, and the recommendation information is sent to the client. By implementing the method, the content possibly interested by the user can be analyzed by combining the historical statistical data, so that more accurate portraits are carried out on the user, and the accuracy of the recommendation result is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 2 is a flow chart of another information recommendation method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating interaction between a client and a server according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an information recommendation device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of an information recommendation method according to an embodiment of the present invention may include:
s101, the server analyzes the search information input by the user to obtain target reference information matched with the search information.
In the embodiment of the invention, the user can input the search information in the preset interface provided by the client, wherein the search information can be a question or a keyword which the user wants to ask, and the question is a question such as "where is the address of a company? The keyword, such as "xiaozhao robot", may be a text input, a voice input, etc., and the manner of inputting the search information by the user is not limited herein.
After receiving the search information input by the user, the client sends the search information to the server, and the server analyzes the search information to obtain target reference information matched with the search information. The similarity between the target reference information and the search information input by the user is greater than a preset similarity threshold, the target reference information can be a preset question, and an answer of the preset question can be obtained by inquiring in a database. In a specific implementation, a plurality of pieces of reference information are stored in the database, each piece of reference information corresponds to one piece of content, and when the reference information is a question, the content corresponding to the reference information is an answer to the question. After obtaining the search information input by the user, the server detects the similarity between each piece of reference information stored in the database and the search information, determines the reference information with the maximum similarity from the detection database, if the similarity between the reference information with the maximum similarity and the search information is larger than a preset similarity threshold value, determines the reference information with the maximum similarity as target reference information, finds the content corresponding to the target reference information from the database, and sends the content to the client for display. It should be noted that, the similarity between the reference information and the search information may be calculated by the server obtaining the number of the same characters in the reference information and the search information, and determining the ratio of the number of the same characters to the total number of the characters in the search information as the similarity between the reference information and the search information. The preset similarity threshold may be 30%, 50%, 70%, etc., and may be specifically preset by a developer.
For example, the search information input by the user is "where the address of a company is", and the reference information stored in the database and the similarity between the calculated reference information and the search information are shown in table 1:
table 1:
| reference information | Content corresponding to reference information | Similarity degree |
| The address of a company is located at | Chongqing | 75％ |
| Personnel count of a company | 200 | 50％ |
| Payroll level of a company | Average 6000 | 50％ |
If the preset similarity threshold is 60%, the server determines that the reference information 'the address of a certain company is located' is the target reference information, and sends the corresponding content 'Chongqing' to the client for display.
S102, the server acquires the historical call times of the target reference information.
In the embodiment of the invention, after determining the target reference information corresponding to the search information input by the user, the server acquires the history call times corresponding to the target reference information, wherein the history call times are times of matching the target reference information with at least one search information input by the user in a first preset time period, that is, times of sending content corresponding to the target reference information to the client by the server in the first preset time period, wherein the first preset time period can be about one year, about one month, about one week, and the like, and can be specifically preset by a researcher.
For example, the preset time period is about one month, the reference information is "the address of a company is located", the server obtains the reference information "the address of a company is located" in about one month as the target reference information, and the number of times of matching the search information input by at least one user is 20, and then the server determines the 20 times as the history call number.
S103, the server determines recommendation information according to the historical call times.
In the embodiment of the invention, after the server acquires the historical call times of the target reference information, the recommendation information is determined according to the historical call times.
Specifically, the server judges whether the historical call times of the target reference information are smaller than the first preset times, if the historical call times of the target reference information are smaller than the first preset times, the intersection length of each piece of reference information stored in the database and the target reference information is obtained, and the reference information with larger intersection length in the database is used as the recommendation information, wherein at least one piece of reference information is stored in the database. In one implementation, the intersection length may be the length of the same character between the reference information and the target reference information in the database. In one implementation manner, the specific determination manner of the intersection length may also be that the server acquires a first core phrase of the first reference information and a second core phrase of the target reference information, where the first reference information is any one of at least one reference information stored in the database, the first core phrase includes at least one first core word, the second core phrase includes at least one second core word, the part of speech of the first core word or the second core word is a preset part of speech, the server acquires the number of the first core words identical to the second core word, and determines the number as the intersection length of the first reference information and the target reference information. The parts of speech may be entity nouns, verbs, adjectives, etc. The first core word and the second core word may be word segmentation processing is performed on each reference information when constructing the database, and the obtained word with the preset part of speech is determined to be the core word. For example, the preset part of speech is an entity noun, the first reference information is "wage level of a company", the first core phrase is "company, wage", the target reference information is "address of a company", the second core phrase is "company, address", and the server determines that the number of first core words identical to the second core words in the first reference information is 1, that is, the intersection length of the first reference information and the target reference information is 1. Alternatively, the core word may be preset for each piece of reference information by a developer when constructing the database.
It should be noted that, the specific determination manner of the reference information with a larger intersection length in the database may be to sort at least one reference information stored in the database according to the order of the intersection length, and determine the reference information sorted into the first n bits as the reference information with a larger intersection length, where n is an integer greater than or equal to 1, and may be specifically preset by a developer or a user.
S104, the server sends recommendation information to the client.
In the embodiment of the invention, after the server determines the recommendation information, the recommendation information is sent to the client and receives the selection operation of the user on the recommendation information, and after the server detects the selection operation of the user on the recommendation information, the server continuously sends the content corresponding to the recommendation information to the client, wherein the recommendation information can be a question which the user hopes to ask, and the content corresponding to the recommendation information is an answer to the question.
As shown in fig. 3, a user inputs search information "where the geographic location of a company is" in a client 301, a server 302 queries that target reference information matched with the search information is "where the geographic location of a company is located", and queries that content corresponding to the target reference information is "Chongqing" in a database, and after the server 302 sends the Chongqing to the client 301 to display, the server also sends recommendation information "the personnel number of a company, and the wage level of a company" to the client 301 to display, so that the user can select the recommendation information.
In the embodiment of the invention, a user inputs search information in a client, the client sends the search information to a server, the server analyzes the search information input by the user to obtain target reference information matched with the search information, the server detects through a core word to find recommendation information similar to the target reference information, and the recommendation information is sent to the client. By implementing the method, the content possibly interested by the user can be analyzed by combining the historical statistical data, so that more accurate portraits are carried out on the user, and the accuracy of the recommendation result is improved.
Referring to fig. 2, a flowchart of another information recommendation method according to an embodiment of the present invention may include:
s201, the server analyzes the search information input by the user to obtain target reference information matched with the search information.
S202, the server acquires the historical call times of the target reference information.
S203, if the historical call times of the target reference information are greater than or equal to the first preset times, the server acquires the session termination times of the target reference information in the first preset time period.
In the embodiment of the invention, after the server obtains the historical call times of the target reference information, whether the historical call times of the target reference information are smaller than the first preset times or not is judged, if the historical call times of the target reference information are larger than or equal to the first preset times, the server obtains the session termination times of the target reference information in a first preset time period, wherein the first preset time period can be the last year, the last month, the last week and the like, the session termination times are the times that after the server sends the content corresponding to the target reference information to the client, the server does not receive the retrieval information input by at least one user again in a second preset time period, and the second preset time period can be 1 hour, 3 hours, one day and the like after the server sends the content corresponding to the target reference information to the client, and can be preset by a research staff.
For example, the preset time period is 1 hour after the server sends the content corresponding to the target reference information to the client, if the search information input by the user is not received again within the hour, the server adds 1 to the session termination frequency of the target reference information recorded in the database, so that the session termination frequency of the target reference information stored in the database is updated.
S204, the server calculates a target ratio between the session termination times and the historical call times of the target reference information, and judges whether the target ratio is larger than a preset ratio.
In the embodiment of the invention, after the server acquires the session termination times about the target reference information in the first preset time period, a target ratio between the session termination times and the historical call times of the target reference information is calculated, and whether the target ratio is larger than the preset ratio is judged. If the target ratio is greater than the preset ratio, step S205 is performed, and if the target ratio is less than or equal to the preset ratio, step S206 is performed.
S205, if the target ratio is larger than the preset ratio, the server pauses outputting the recommended information.
In the embodiment of the invention, if the target ratio between the session termination times and the historical call times of the target reference information is larger than the preset ratio, the target reference information is often matched with the last retrieval information input by at least one user in one session, and the probability that the user does not retrieve other information after acquiring the content corresponding to the target recommendation information is larger. And the server pauses outputting the recommended information after sending the content corresponding to the target reference information to the client, namely, does not output the corresponding recommended information aiming at the search information input by the user, and finishes the flow.
S206, if the target ratio is smaller than or equal to the preset ratio, the server determines recommended information according to the association times between the reference information and the target reference information.
In the embodiment of the invention, after the server determines that the target ratio between the session termination times and the historical call times of the target reference information is smaller than or equal to the preset ratio, the server determines the recommended information according to the association times between the reference information stored in the database and the target reference information.
In one implementation manner, the server judges whether the historical call number of the target reference information is greater than the second preset number of times, if the historical call number of the target reference information is greater than the second preset number of times, the association number of times between each reference information stored in the database in a third preset time period and the target reference information is obtained, wherein the second preset number of times is greater than the first preset number of times, the association number of times is the number of times that the server receives the reference information input by the user after outputting the content corresponding to the target reference information, the third preset time period can be the last year, the last month, the last week, etc., the second preset time period can be 1 hour, 3 hours, one day, etc., after the server sends the content corresponding to the target reference information to the client, the specific construction manner of the association number of each reference information in the database can be that the server constructs an association feature matrix M according to the search information received by the history and n reference information stored in the database in advance, wherein M is a matrix of n x n, M (i, j) represents the number of times that the index information is input by the user in the first preset time period as the reference information, and the reference information is matched with the reference information input by the user in the second preset time period. After receiving the new search information input by the user, the server updates the stored association feature matrix according to the search information input by the user.
After the server obtains the association times between each piece of reference information and the target reference information stored in the database in the third preset time period, the reference information with larger association times can be directly determined as recommended information, in one implementation manner, the specific determination manner of the reference information with larger association times can be that the server ranks at least one piece of reference information stored in the database according to the order of magnitude of association times, and the reference information ranked as the first n bits is determined as the reference information with larger association times, wherein n is an integer greater than or equal to 1, and the specific determination manner can be preset by a research and development personnel or a user. In one implementation manner, after the server obtains the reference information with the association value ordered into the first n bits, it will also detect whether the association number of the reference information ordered into the first n bits is greater than a preset association number, if yes, the reference information ordered into the first n bits is determined to be the reference information with the larger association number, where the preset association number may be an average value, a double average value, or the like of association numbers of each reference information stored in the database. If the association times of the reference information sequenced to the first n bits are smaller than the preset times, the server sequences at least one reference information stored in the database according to the sequence of the intersection length of the reference information and the target reference information, and determines the reference information sequenced to the first n bits as recommended information.
In one implementation manner, if the historical call number of times of the target reference information is less than or equal to the second preset number of times, the server obtains an intersection length between each reference information and the target reference information stored in the database in a third preset time period and an association number of times between each reference information and the target reference information, and processes the intersection length and the association number of times through a preset algorithm to obtain an association value, wherein a specific determination algorithm of the association value may be:
Y(i，j)＝log(M(i，j))*T(i，j)
wherein Y (i, j) represents an association value between the reference information j and the target reference information i, M (i, j) represents the association number between the reference information j and the target reference information i, T (i, j) represents the intersection length between the reference information j and the target reference information i, and the server determines the reference information with a larger association value as the recommendation information. The reference information with a larger association value may be reference information with the association value ordered as the first n bits.
S207, the server sends recommendation information to the client.
In one implementation manner, before the server sends the recommended information to the client, the server further screens the recommended information, specifically, the server obtains a target historical time for sending the recommended information to the client last time, detects whether a time interval between the target historical time and a current system time is greater than a preset time interval, and if so, executes an operation for sending the recommended information to the client. By the method, the server can be prevented from repeatedly recommending the content which is not interested by the user and is recently recommended.
In one implementation manner, after determining the recommendation information, the server may directly send the recommendation information to the client and receive a selection operation of the user on the recommendation information, and after detecting the selection operation of the user on the recommendation information, the server will continue to send content corresponding to the recommendation information to the client, where the recommendation information may be a question that the user wishes to ask, and the content corresponding to the recommendation information is an answer to the question.
In the embodiment of the invention, the user inputs the information to be searched in the client, the client sends the search information to the server, the server analyzes the search information input by the user to obtain the target reference information matched with the search information and the historical call times of the target reference information, if the historical call times of the target reference information are greater than or equal to the first preset times, the server acquires the session termination times of the target reference information in the first preset time period, and for the target reference information with the greater session termination times, the server pauses outputting the recommended information, thereby reducing the workload of the server, and for the target reference information with the smaller session termination times, the recommended information is determined continuously through the intersection length or the association times between the reference information and the target reference information in the database.
The following details the content of embodiments 1-2 of the present invention in conjunction with a specific application scenario, and assume that n sets of reference information are stored in the database, the search information input by the user is user_ que, and the reference information que _i already entered in the database is matched, that is, que _i is the target reference information. The server extracts information association features M (i, i) of the users, recently queried database information features R, association core word features T, end features E and matching times features F from the historical session conditions of the users stored in the database.
Wherein, the information association feature M (i) is a matrix of n, M (i, j), j=1..n represents the frequency of matching the reference information que _j in the database after the search information input by the user matches the reference information que _i in the database. The most recently interrogated database information feature R is a row vector of length n, R (i) representing that the reference information que _i in the database is matched by the user in the previous R (i) dialogs. If r (i) is 0 for the current dialogue, if que _i is not matched all the time, taking the set maximum value, taking the associated core word feature T as a row vector with the length of n, wherein T (j) represents the intersection length of the core word corresponding to the target reference information and the core word of the reference information que _j in the database, and the ending feature E represents the number of times of ending the dialogue after the search information input by the user is matched with the reference information que _i in the database, wherein E is a row vector. The matching number feature F indicates the number of times the reference information que _i is matched with the search information input by the user, and F is a row vector.
Assuming that the set of recommendation information is C, the number of recommendation information is MaxC.
If F (i) < f_thred0, it indicates that the reference information que _i in the database is rarely asked for, at which time c=rule_t (T). Wherein F (i) represents the number of times that the reference information que _i in the database matches the search information input by the user, f_thred0 represents a first preset threshold value, which can be preset by a developer, rule_t (T) represents that the reference information is ordered from high to low according to a T value, reference information of the prior MaxC is selected to form a set C of recommended information, and the T value represents the intersection length of the core word corresponding to the target reference information and the core word of the reference information in the database.
If F (i) > thread0 and E (i)/F (i) > ef_thread, then no recommendation is made. At this time: c=empty set. Wherein, E (i) represents the number of times the session ends after the search information input by the user matches the reference information que _i in the database, and ef_thread represents a preset ratio, which can be preset by a developer.
If F (i) > F_thred0 and E (i)/F (i) < ef_thred0 and F (i) > F_thred1, then it is indicated that the reference information que _i in the database is always asked for, at this time
1) C=rule_m (M (i):), the reference information is ordered from high to low according to the value of M (i): and the reference information of the previous MaxC is selected as C.
2) And (3) filtering: que j belongs to C, satisfies M (i, j) >2 x sum (M (i,:)/n), and requires more than 2 times the average correlation value to be recommended.
3) If C is an empty set after filtering, then c=rule_t (T). Wherein, thread1 represents a second preset threshold, thread1> thread0.
If F (i) > F_thred0 and E/F (i) < ef_thred1 and F (i) < F_thred1:
c=rule_mt (M (i, i): T), i.e. the reference information of the previous MaxC is selected as C, ordered from high to low in terms of log (M (i):)) T value.
Filtering the reference information appearing in the last R_thread time in the current session in C by using the result in the R filter C, and:
C＝C-R(R(j)<R_thread)。
after determining the set C of recommendation information, the server sends the set C to the client of the user.
The information recommending apparatus according to the embodiment of the present invention will be described in detail with reference to fig. 4. It should be noted that, the information recommending apparatus shown in fig. 4 is used to execute the method of the embodiment shown in fig. 1-2, and for convenience of explanation, only the portion relevant to the embodiment of the present invention is shown, and specific technical details are not disclosed, and reference is made to the embodiment shown in fig. 1-2 of the present invention.
Referring to fig. 4, a schematic structural diagram of an information recommendation device provided by the present invention, the information recommendation device 40 may include: an analysis module 401, an acquisition module 402, a determination module 403, and a transmission module 404.
The analysis module 401 is configured to analyze the search information input by the user to obtain target reference information matched with the search information, where a similarity between the search information and the target reference information is greater than a preset similarity threshold;
an obtaining module 402, configured to obtain a history call number of times of the target reference information, where the history call number of times is a number of times that the target reference information matches with at least one search information input by a user in a first preset period of time;
a determining module 403, configured to determine recommendation information according to the historical call times;
and the sending module 404 is configured to send the recommendation information to the client.
In one implementation, the determining module 403 is specifically configured to:
judging whether the historical calling times of the target reference information are smaller than a first preset times or not;
if the historical call times of the target reference information are smaller than the first preset times, acquiring the intersection length of each piece of reference information stored in a database and the target reference information, wherein the database comprises at least one piece of reference information;
and determining the reference information with larger intersection length in the database as the recommendation information.
In one implementation, the determining module 403 is specifically configured to:
the method comprises the steps of obtaining a first core phrase of first reference information and a second core phrase of target reference information, wherein the first reference information is any one of at least one piece of reference information stored in a database, the first core phrase comprises at least one first core word, the second core phrase comprises at least one second core word, and the part of speech of the first core word or the second core word is a preset part of speech;
acquiring the number of first core words identical to the second core words;
the number is determined as an intersection length of the first reference information and the target reference information.
In one implementation, the determining module 403 is specifically configured to:
if the historical calling times of the target reference information are greater than or equal to a first preset times, acquiring session termination times of the target reference information in a first preset time period, wherein the session termination times are times when user input retrieval information is not received again in a second preset time period after content corresponding to the target reference information is sent to a client;
Calculating a target ratio between the session termination times and the historical call times of the target reference information;
and if the target ratio is larger than the preset ratio, suspending outputting the recommended information.
In one implementation, the determining module 403 is specifically configured to:
if the target ratio is smaller than or equal to a preset ratio, judging whether the historical calling times of the target reference information are larger than a second preset times or not, wherein the second preset times are larger than the first preset times;
if the historical calling times of the target reference information are larger than the second preset times, acquiring the association times between each piece of reference information stored in the database and the target reference information in a third preset time period, wherein the association times are the times of receiving the reference information input by a user in the second preset time period after outputting the content corresponding to the target reference information;
and determining the reference information with larger association times as recommendation information.
In one implementation, the determining module 403 is specifically configured to:
if the historical calling times of the target reference information are smaller than or equal to the second preset times, acquiring the intersection length between each piece of reference information stored in the database and the target reference information and the association times between each piece of reference information and the target reference information in a third preset time period;
Processing the intersection length and the association times through a preset algorithm to obtain an association value;
and determining the reference information with a larger association value as recommendation information.
In one implementation, the determining module 403 is further configured to:
acquiring target historical time for sending the recommendation information to the client in the last time;
detecting whether the time interval between the target historical time and the current system time is larger than a preset time interval or not;
if yes, the operation of sending the recommendation information to the client is executed.
In the embodiment of the invention, a user inputs search information in a client, the client sends the search information to a server, an analysis module 401 analyzes the search information input by the user to obtain target reference information matched with the search information, the similarity between the search information and the target reference information is greater than a preset similarity threshold, and an acquisition module 402 acquires the historical call times of the target reference information, wherein the historical call times are times of matching the target reference information with at least one search information input by the user in a first preset time period; the determining module 403 determines recommendation information according to the historical call times; the sending module 404 sends the recommendation information to the client. By implementing the method, the content possibly interested by the user can be analyzed by combining the historical statistical data, so that more accurate portraits are carried out on the user, and the accuracy of the recommendation result is improved.
Referring to fig. 5, a schematic structural diagram of a server is provided in an embodiment of the present invention. As shown in fig. 5, the server includes: at least one processor 501, an input device 503, an output device 504, a memory 505, and at least one communication bus 502. Wherein a communication bus 502 is used to enable connected communications between these components. The input device 503 may be a control panel, a microphone, or the like, and the output device 504 may be a display screen or the like. The memory 505 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The memory 505 may also optionally be at least one storage device located remotely from the processor 501. Wherein the processor 501 may be described in connection with fig. 4 as an information recommendation apparatus, a set of program codes is stored in the memory 505, and the processor 501, the input device 503, and the output device 504 call the program codes stored in the memory 505 for performing the following operations:
the processor 501 is configured to analyze search information input by a user to obtain target reference information matched with the search information, where a similarity between the search information and the target reference information is greater than a preset similarity threshold;
The processor 501 is configured to obtain a history call number of times of the target reference information, where the history call number of times is a number of times that the target reference information matches with at least one search information input by a user in a first preset time period;
a processor 501 for determining recommendation information according to the historical call times;
and an output device 504, configured to send the recommendation information to the client.
In one implementation, the processor 501 is specifically configured to:
judging whether the historical calling times of the target reference information are smaller than a first preset times or not;
if the historical call times of the target reference information are smaller than the first preset times, acquiring the intersection length of each piece of reference information stored in a database and the target reference information, wherein the database comprises at least one piece of reference information;
and determining the reference information with larger intersection length in the database as the recommendation information.
In one implementation, the processor 501 is specifically configured to:
the method comprises the steps of obtaining a first core phrase of first reference information and a second core phrase of target reference information, wherein the first reference information is any one of at least one piece of reference information stored in a database, the first core phrase comprises at least one first core word, the second core phrase comprises at least one second core word, and the part of speech of the first core word or the second core word is a preset part of speech;
Acquiring the number of first core words identical to the second core words;
the number is determined as an intersection length of the first reference information and the target reference information.
In one implementation, the processor 501 is specifically configured to:
if the historical calling times of the target reference information are greater than or equal to a first preset times, acquiring session termination times of the target reference information in a first preset time period, wherein the session termination times are times when user input retrieval information is not received again in a second preset time period after content corresponding to the target reference information is sent to a client;
calculating a target ratio between the session termination times and the historical call times of the target reference information;
and if the target ratio is larger than the preset ratio, suspending outputting the recommended information.
In one implementation, the processor 501 is specifically configured to:
if the target ratio is smaller than or equal to a preset ratio, judging whether the historical calling times of the target reference information are larger than a second preset times or not, wherein the second preset times are larger than the first preset times;
if the historical calling times of the target reference information are larger than the second preset times, acquiring the association times between each piece of reference information stored in the database and the target reference information in a third preset time period, wherein the association times are the times of receiving the reference information input by a user in the second preset time period after outputting the content corresponding to the target reference information;
And determining the reference information with larger association times as recommendation information.
In one implementation, the processor 501 is specifically configured to:
if the historical calling times of the target reference information are smaller than or equal to the second preset times, acquiring the intersection length between each piece of reference information stored in the database and the target reference information and the association times between each piece of reference information and the target reference information in a third preset time period;
processing the intersection length and the association times through a preset algorithm to obtain an association value;
and determining the reference information with a larger association value as recommendation information.
In one implementation, the processor 501 is specifically configured to:
acquiring target historical time for sending the recommendation information to the client in the last time;
detecting whether the time interval between the target historical time and the current system time is larger than a preset time interval or not;
if yes, the operation of sending the recommendation information to the client is executed.
In the embodiment of the invention, the processor 501 analyzes the search information input by the user to obtain the target reference information matched with the search information, and the similarity between the search information and the target reference information is greater than a preset similarity threshold; the processor 501 obtains the historical call times of the target reference information, wherein the historical call times are times of matching the target reference information with at least one search information input by a user in a first preset time period; the processor 501 determines recommendation information according to the historical call times; the output device 504 sends the recommendation information to the client. The content which is possibly interested by the user can be analyzed by combining the historical statistical data, so that more accurate portrayal is carried out on the user, and the accuracy of the recommendation result is improved.
The modules described in the embodiments of the present invention may be implemented by general-purpose integrated circuits such as a CPU (Central Processing Unit ) or by ASIC (Application Specific Integrated Circuit, application specific integrated circuit).
It should be appreciated that in embodiments of the present invention, the processor 501 may be a central processing module (Central Processing Unit, CPU), which may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSPs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The bus 502 can be an industry standard architecture (Industry Standard Architecture, ISA) bus, an external device interconnect (Peripheral Component, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, etc., and the bus 502 can be divided into an address bus, a data bus, a control bus, etc., with fig. 5 being shown with only one bold line for ease of illustration, but not with only one bus or one type of bus.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in the embodiments may be accomplished by way of a computer program stored in a computer storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The computer storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), or the like.
The foregoing disclosure is illustrative of the present invention and is not to be construed as limiting the scope of the invention, which is defined by the appended claims.
Claims (9)
1. An information recommendation method, the method comprising:
analyzing the search information input by a user to obtain target reference information matched with the search information, wherein the similarity between the search information and the target reference information is larger than a preset similarity threshold;
acquiring the historical call times of the target reference information, wherein the historical call times are times of matching of the target reference information with at least one search information input by a user in a first preset time period;
If the historical call times are smaller than the first preset times, acquiring the intersection length of each piece of reference information stored in a database and the target reference information, wherein the database comprises at least one piece of reference information;
ordering at least one reference information stored in the database according to the order of the intersection length;
determining reference information sequenced into the first n bits as recommended information, wherein n is an integer greater than or equal to 1;
and sending the recommendation information to the client.
2. The method of claim 1, wherein the acquiring the intersection length of each reference information stored in the database with the target reference information comprises:
the method comprises the steps of obtaining a first core phrase of first reference information and a second core phrase of target reference information, wherein the first reference information is any one of at least one piece of reference information stored in a database, the first core phrase comprises at least one first core word, the second core phrase comprises at least one second core word, and the part of speech of the first core word or the second core word is a preset part of speech;
acquiring the number of first core words identical to the second core words;
The number is determined as an intersection length of the first reference information and the target reference information.
3. The method according to claim 1, wherein the method further comprises:
if the historical calling times of the target reference information are greater than or equal to a first preset times, acquiring session termination times of the target reference information in a first preset time period, wherein the session termination times are times when user input retrieval information is not received again in a second preset time period after content corresponding to the target reference information is sent to a client;
calculating a target ratio between the session termination times and the historical call times of the target reference information;
and if the target ratio is larger than the preset ratio, suspending outputting the recommended information.
4. The method of claim 3, further comprising, after said calculating a target ratio between said session termination number and said historical call number of said target reference information:
if the target ratio is smaller than or equal to a preset ratio, judging whether the historical calling times of the target reference information are larger than a second preset times or not, wherein the second preset times are larger than the first preset times;
If the historical calling times of the target reference information are larger than the second preset times, acquiring the association times between each piece of reference information stored in the database and the target reference information in a third preset time period, wherein the association times are the times of receiving the reference information input by a user in the second preset time period after outputting the content corresponding to the target reference information;
sequencing at least one piece of reference information stored in the database according to the order of the association times;
and determining the reference information sequenced into the first n bits as the recommendation information, wherein n is an integer greater than or equal to 1.
5. The method of claim 4, wherein after determining whether the historical call count of the target reference information is greater than a second preset count, further comprising:
if the historical calling times of the target reference information are smaller than or equal to the second preset times, acquiring the intersection length between each piece of reference information stored in the database and the target reference information and the association times between each piece of reference information and the target reference information in a third preset time period;
Processing the intersection length and the association times through a preset algorithm to obtain an association value;
sequencing at least one piece of reference information stored in the database according to the magnitude sequence of the association values;
and determining the reference information sequenced into the first n bits as the recommendation information, wherein n is an integer greater than or equal to 1.
6. The method of claim 1, wherein after determining recommendation information based on the historical call count, further comprising:
acquiring target historical time for sending the recommendation information to the client in the last time;
detecting whether the time interval between the target historical time and the current system time is larger than a preset time interval or not;
if yes, the operation of sending the recommendation information to the client is executed.
7. An information recommendation device, characterized by comprising:
the analysis module is used for analyzing the search information input by the user to obtain target reference information matched with the search information, and the similarity between the search information and the target reference information is larger than a preset similarity threshold;
the acquisition module is used for acquiring the historical call times of the target reference information, wherein the historical call times are times of matching of the target reference information and at least one search information input by a user in a first preset time period;
The determining module is used for acquiring the intersection length of each piece of reference information stored in a database and the target reference information if the historical call times are smaller than the first preset times, wherein the database comprises at least one piece of reference information; ordering at least one reference information stored in the database according to the order of the intersection length; determining reference information sequenced into the first n bits as recommended information, wherein n is an integer greater than or equal to 1;
and the sending module is used for sending the recommendation information to the client.
8. A server comprising a processor and a memory, the processor and the memory being interconnected, wherein the memory is adapted to store a computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method of any of claims 1-6.
9. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program comprising program instructions which, when executed by a processor, cause the processor to perform the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811238927.3A CN109597874B (en) | 2018-10-23 | 2018-10-23 | Information recommendation method, device and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811238927.3A CN109597874B (en) | 2018-10-23 | 2018-10-23 | Information recommendation method, device and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109597874A CN109597874A (en) | 2019-04-09 |
| CN109597874B true CN109597874B (en) | 2023-05-23 |
Family
ID=65957963
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811238927.3A Active CN109597874B (en) | 2018-10-23 | 2018-10-23 | Information recommendation method, device and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109597874B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112733006B (en) * | 2019-10-14 | 2022-12-02 | 中国移动通信集团上海有限公司 | User portrait generation method, device and equipment and storage medium |
| CN110855483A (en) * | 2019-11-05 | 2020-02-28 | 杭州安恒信息技术股份有限公司 | Operation and maintenance command management method and system and electronic equipment |
| CN111046286B (en) * | 2019-12-12 | 2023-04-18 | 腾讯科技（深圳）有限公司 | Object recommendation method and device and computer storage medium |
| CN113472856A (en) * | 2021-06-08 | 2021-10-01 | 北京城市网邻信息技术有限公司 | Message pushing processing method and device, electronic equipment and storage medium |
| CN113469704A (en) * | 2021-06-10 | 2021-10-01 | 云南电网有限责任公司 | System and method for automatically recommending customer service strategy |
| CN113468279B (en) * | 2021-06-30 | 2023-09-08 | 杭州云深科技有限公司 | System for acquiring association relation of target users |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101465749A (en) * | 2008-12-29 | 2009-06-24 | 武汉大学 | Method for building interlocution service based on Web Service combination |
| CN106095988A (en) * | 2016-06-21 | 2016-11-09 | 上海智臻智能网络科技股份有限公司 | Automatic question-answering method and device |
| CN107402961A (en) * | 2017-06-16 | 2017-11-28 | 北京三快在线科技有限公司 | One kind recommends method and device, electronic equipment |
| CN107609056A (en) * | 2017-08-25 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | A kind of question and answer processing method and equipment based on picture recognition |
| CN107784092A (en) * | 2017-10-11 | 2018-03-09 | 深圳市金立通信设备有限公司 | A kind of method, server and computer-readable medium for recommending hot word |
| CN108681564A (en) * | 2018-04-28 | 2018-10-19 | 北京京东尚科信息技术有限公司 | The determination method, apparatus and computer readable storage medium of keyword and answer |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9251360B2 (en) * | 2012-04-27 | 2016-02-02 | Intralinks, Inc. | Computerized method and system for managing secure mobile device content viewing in a networked secure collaborative exchange environment |
-
2018
- 2018-10-23 CN CN201811238927.3A patent/CN109597874B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101465749A (en) * | 2008-12-29 | 2009-06-24 | 武汉大学 | Method for building interlocution service based on Web Service combination |
| CN106095988A (en) * | 2016-06-21 | 2016-11-09 | 上海智臻智能网络科技股份有限公司 | Automatic question-answering method and device |
| CN107402961A (en) * | 2017-06-16 | 2017-11-28 | 北京三快在线科技有限公司 | One kind recommends method and device, electronic equipment |
| CN107609056A (en) * | 2017-08-25 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | A kind of question and answer processing method and equipment based on picture recognition |
| CN107784092A (en) * | 2017-10-11 | 2018-03-09 | 深圳市金立通信设备有限公司 | A kind of method, server and computer-readable medium for recommending hot word |
| CN108681564A (en) * | 2018-04-28 | 2018-10-19 | 北京京东尚科信息技术有限公司 | The determination method, apparatus and computer readable storage medium of keyword and answer |
Non-Patent Citations (3)
| Title |
|---|
| A similarity search method of time series data with combination of Fourier and wavelet transforms;K. Kawagoe et al.;《Proceedings Ninth International Symposium on Temporal Representation and Reasoning》;20021107;86-92 * |
| 基于社区问答系统的探索式搜索查询扩展方法研究;高礼;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170315;I138-5968 * |
| 智能推荐引擎中心管理系统的设计与实现;范艳华;《有线电视技术》;20160131(第313期);106-108 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109597874A (en) | 2019-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109597874B (en) | Information recommendation method, device and server | |
| CN107609101B (en) | Intelligent interaction method, equipment and storage medium | |
| CN106874441B (en) | Intelligent question-answering method and device | |
| WO2017097061A1 (en) | Smart response method and apparatus | |
| CN109167816B (en) | Information pushing method, device, equipment and storage medium | |
| CN110069698B (en) | Information pushing method and device | |
| CN107526846B (en) | Method, device, server and medium for generating and sorting channel sorting model | |
| US20130006975A1 (en) | System and method for matching entities and synonym group organizer used therein | |
| US20190114711A1 (en) | Financial analysis system and method for unstructured text data | |
| CN112487150B (en) | File management method, system, storage medium and electronic equipment | |
| CN109977233B (en) | Idiom knowledge graph construction method and device | |
| CN109522397B (en) | Information processing method and device | |
| CN112199588A (en) | Public opinion text screening method and device | |
| CN110941702A (en) | Retrieval method and device for laws and regulations and laws and readable storage medium | |
| KR102285232B1 (en) | Morphology-Based AI Chatbot and Method How to determine the degree of sentence | |
| CN109344232B (en) | Public opinion information retrieval method and terminal equipment | |
| CN111435406A (en) | Method and device for correcting database statement spelling errors | |
| CN113342954A (en) | Image information processing method and device applied to question-answering system and electronic equipment | |
| CN110837732A (en) | Method and device for identifying intimacy between target people, electronic equipment and storage medium | |
| KR101923820B1 (en) | Personal customized sentence automatic recommendation foreign language learning system | |
| CN105893397A (en) | Video recommendation method and apparatus | |
| CN115186063A (en) | Service data processing method, device, equipment and storage medium | |
| KR102308521B1 (en) | Method and device for updating information | |
| CN110895924B (en) | Method and device for reading document content aloud, electronic equipment and readable storage medium | |
| CN112785095A (en) | Loan prediction method, loan prediction device, electronic device, and computer-readable 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 |