CN113065067A - Article recommendation method and device, computer equipment and storage medium - Google Patents
Article recommendation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113065067A CN113065067A CN202110351648.3A CN202110351648A CN113065067A CN 113065067 A CN113065067 A CN 113065067A CN 202110351648 A CN202110351648 A CN 202110351648A CN 113065067 A CN113065067 A CN 113065067A
- Authority
- CN
- China
- Prior art keywords
- user
- article
- vector
- recommended
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 239000013598 vector Substances 0.000 claims abstract description 144
- 230000006399 behavior Effects 0.000 claims description 173
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 238000012163 sequencing technique Methods 0.000 claims description 8
- 238000010801 machine learning Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 230000003542 behavioural effect Effects 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 238000011478 gradient descent method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses an article recommendation method, an article recommendation device, computer equipment and a storage medium. The method comprises the following steps: the method comprises the steps of obtaining at least one effective behavior data matched with a target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence; determining an item vector of at least one item to be recommended; and determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user. The scheme of the embodiment of the invention solves the problems of low accuracy and poor effect of recommending the articles to the user, and realizes that the articles of interest are accurately recommended to the user.
Description
Technical Field
The embodiment of the invention relates to the technical field of information processing, in particular to an article recommendation method and device, computer equipment and a storage medium.
Background
With the continuous development and popularization of information technology, information resources on the internet grow exponentially, and massive information resources have various characteristics of isomerism, multivariate, distribution and the like. Therefore, according to the information of the user, the interested content is recommended to the user in a targeted manner, so that personalized service provided for the user is widely researched; the content of interest to the user is an item, such as a novel, a commodity, a small video, or news.
At present, the characteristics of the article or the user are constructed mainly according to the label of the article or the user, for example, the label of the article category, the keyword, the user gender, the age or the region, and the like, and then the article is recommended to the user.
In the method in the prior art, the difficulty in obtaining the labels of articles or users is high, the labels of the users such as gender, age or region need to be actively provided by the users, the privacy awareness of the users is stronger and stronger at present, and most of the users cannot actively provide the information; meanwhile, the label of the article or the user needs to be edited and generated manually, so that the cost is high, the quality of the label often depends on an editor, the quality of the label is difficult to control, and the accuracy rate of recommending the article to the user is low and the effect is poor.
Disclosure of Invention
The embodiment of the invention provides an article recommendation method, an article recommendation device, computer equipment and a storage medium, which are used for accurately recommending interested articles to a user.
In a first aspect, an embodiment of the present invention provides an item recommendation method, including:
the method comprises the steps of obtaining at least one effective behavior data matched with a target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence;
determining an item vector of at least one item to be recommended;
and determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
In a second aspect, an embodiment of the present invention further provides an article recommendation apparatus, including:
the user vector determining module is used for acquiring at least one effective behavior data matched with a target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence;
the article vector determining module is used for determining an article vector of at least one article to be recommended;
and the target recommended article determining module is used for determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the item recommendation method according to any embodiment of the present invention.
In a fourth aspect, the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are used to execute the item recommendation method according to any one of the embodiments of the present invention when executed by a computer processor.
The embodiment of the invention divides each effective behavior data by acquiring at least one effective behavior data matched with a target user to obtain a plurality of user behavior sequences, and determines a user vector corresponding to the target user according to each user behavior sequence; determining an item vector of at least one item to be recommended; the target recommended article is determined according to the article vector of each article to be recommended and the user vector, and the target recommended article is displayed to the user, so that the problems of low accuracy and poor effect of recommending the article to the user are solved, and the article of interest is accurately recommended to the user.
Drawings
FIG. 1 is a flow chart of a method for recommending items according to a first embodiment of the present invention;
FIG. 2 is a flowchart of an item recommendation method according to a second embodiment of the present invention;
FIG. 3 is a flowchart of an item recommendation method according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an article recommendation device according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of a computer device according to a fifth embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in further detail with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad invention. It should be further noted that, for convenience of description, only some structures, not all structures, relating to the embodiments of the present invention are shown in the drawings.
Example one
Fig. 1 is a flowchart of an item recommendation method in a first embodiment of the present invention, where this embodiment is applicable to a case of recommending an item of interest to a user, and the method may be executed by an item recommendation apparatus, and the apparatus may be implemented by software and/or hardware and integrated in a computer device. Specifically, referring to fig. 1, the method specifically includes the following steps:
and 110, acquiring at least one effective behavior data matched with the target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence.
The effective behavior data can be behaviors of clicking, playing, forwarding and reading a certain article for a user; the articles may be novels of a novels website, various commodities of a shopping platform, or articles of an information platform, which is not limited in this embodiment.
For example, the at least one effective behavior data matched with the target user may be a name or a link of a small article read by the target user at 8 am, attribute information or a link of a product browsed at 6 pm, a name or an author of a novel read at 10 pm, and the like.
In an optional implementation manner of this embodiment, after at least one piece of valid behavior data that matches the target user is obtained, each piece of valid behavior data may be further segmented, for example, each piece of valid behavior data may be segmented according to time, for example, a plurality of pieces of valid behavior data in the same time period may be segmented into one user behavior sequence.
Further, a user vector corresponding to the target user can be determined according to the divided user behavior sequences; illustratively, key features of each user behavior sequence can be extracted, and a user vector corresponding to a target user is generated according to each key feature; it will be appreciated that the user vector corresponding to the target user is associated with the item contained in the valid behavior data that the target user matches.
And step 120, determining an item vector of at least one item to be recommended.
The items to be recommended may be all novels in a novels website, all commodities in a shopping platform, or all articles in an information platform, and the like, which is not limited in this embodiment.
In an optional implementation manner of this embodiment, after obtaining each item to be recommended, an item vector of each item to be recommended may be further determined, where the item vector of each item may include attribute information such as a name, an author, an address, or a size of the item, and this is not limited in this embodiment.
And step 130, determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
The target recommended item may be one item or a plurality of items, which is not limited in this embodiment.
In an optional implementation manner of this embodiment, after determining a user vector corresponding to a target user and an item vector of at least one item to be recommended, a target recommended item may be further determined according to the user vector and each item vector, and the target recommended item may be presented to the user, so that the user may quickly select the target recommended item.
For example, the association degree between the user vector and each item vector may be calculated, and further, the target recommended item may be obtained by screening according to the association degree.
In the embodiment, at least one effective behavior data matched with a target user is obtained, each effective behavior data is segmented to obtain a plurality of user behavior sequences, and a user vector corresponding to the target user is determined according to each user behavior sequence; determining an item vector of at least one item to be recommended; the target recommended article is determined according to the article vector of each article to be recommended and the user vector, and the target recommended article is displayed to the user, so that the problems of low accuracy and poor effect of recommending the article to the user are solved, and the article of interest is accurately recommended to the user.
Example two
Fig. 2 is a flowchart of an article recommendation method in a second embodiment of the present invention, which is a further refinement of the above technical solutions, and the technical solutions in this embodiment may be combined with various alternatives in one or more of the above embodiments. As shown in fig. 2, the item recommendation method may include the steps of:
Wherein the behavioral data includes at least one of: click, play, forward, and read.
In an optional implementation manner of this embodiment, behavior data matched with a target user may be obtained in real time, and the obtained behavior data is screened, so as to obtain effective behavior data; for example, if the behavior data a matched with the target user is the behavior of reading the novel a, if the reading duration is greater than a first set threshold (for example, 30 seconds, 40 seconds, or 50 seconds, etc., which is not limited in this embodiment), the behavior data a may be determined to be valid behavior data; if the behavior data B matched with the target user is a behavior of playing the video a, if the playing time length is greater than a second set threshold (for example, 60 seconds, 70 seconds, or 80 seconds, etc., which is not limited in this embodiment), it may be determined that the behavior data B is valid behavior data.
And step 220, sequencing the effective behavior data according to a preset sequence, and segmenting the effective behavior data according to a preset interval threshold to obtain a plurality of user behavior sequences.
Wherein, the preset sequence can be a time sequence; the preset interval threshold may be 10 seconds, 1 minute, or 10 minutes, and the like, which is not limited in this embodiment.
In an optional implementation manner of this embodiment, after the effective behavior data matched with the target user is obtained, the effective behavior data may be further sorted according to a time sequence, and the sorted effective behavior data is segmented, so as to obtain a plurality of user behavior sequences.
In an optional implementation manner of this embodiment, segmenting each valid behavior data according to a preset interval threshold to obtain a plurality of user behavior sequences may include: recording the target starting time of the last effective behavior data of the user, and calculating the time interval between the effective starting time of the current effective behavior data and the target starting time; and when the time interval is greater than a preset interval threshold, storing each effective behavior data in the time interval into the same behavior sequence.
In a specific example of this embodiment, after sorting the effective behavior data according to the chronological order, the actual start time and the current position information of the effective behavior data may be recorded from the first effective behavior data; traversing subsequent behaviors, and comparing the time interval of the next valid behavior data with the actual starting time of the first valid behavior data; if the time interval is greater than a preset interval threshold (e.g., 30 seconds), the valid behavior data within the time interval are stored in the same user behavior sequence until the traversal ends with all valid behavior data.
In an optional implementation manner of this embodiment, after obtaining a plurality of user behavior sequences, each user behavior sequence may be further input into a pre-trained prediction model, so as to obtain each item vector corresponding to each user behavior sequence; further, each item vector is subjected to accumulation calculation, so that a user vector corresponding to the target user is generated.
In an optional implementation manner of this embodiment, before inputting each user behavior sequence into the pre-trained prediction model, the method may further include: acquiring training behavior sequences matched with a plurality of users; and training the set machine learning model by using the training behavior sequence to obtain a prediction model.
The preset machine learning model may be an item2vec model or an RNN model, which is not limited in this embodiment.
In a specific example of this embodiment, 100000 behavior sequences of 1000 users can be obtained, and the item2vec machine learning model is trained according to the 100000 behavior sequences, so as to obtain the prediction model.
It should be noted that, in the training process, the item vector may be multiplied by the weight matrix w1 to obtain a hidden layer vector; multiplying the hidden layer vector by an output weight matrix w2 to obtain an output vector; and comparing the output vector with the vector of the context article of the article, and updating the weight matrixes w1 and w2 according to a gradient descent method to obtain final model parameters w1 and w 2.
And 240, acquiring each article to be recommended, performing unique hot coding on each article to be recommended respectively, and determining an article vector matched with each article to be recommended.
In an optional implementation manner of the embodiment, each item to be recommended may be subjected to unique hot coding, so that an item vector matching each item to be recommended may be determined, which has an advantage that an item vector of each item with recommendation may be determined quickly, and a basis is provided for subsequently determining a target recommended item.
And step 250, determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
In the scheme of this embodiment, the effective behavior data are sorted according to a preset sequence, and are segmented according to a preset interval threshold to obtain a plurality of user behavior sequences; respectively inputting each user behavior sequence into a pre-trained prediction model to obtain an article vector corresponding to each user behavior sequence; the method and the device perform accumulation calculation on the object vectors to generate the user vectors corresponding to the target users, can quickly determine the user vectors, provide basis for subsequently determining the target recommended objects, and improve the object recommendation efficiency.
EXAMPLE III
Fig. 3 is a flowchart of an article recommendation method in a third embodiment of the present invention, which is a further refinement of the above technical solutions, and the technical solutions in this embodiment may be combined with various alternatives in one or more of the above embodiments. As shown in fig. 3, the item recommendation method may include the steps of:
and 310, acquiring at least one behavior data matched with the target user, and screening each behavior data according to a set threshold value to obtain each effective behavior data.
And 320, sequencing the effective behavior data according to a preset sequence, and segmenting the effective behavior data according to a preset interval threshold to obtain a plurality of user behavior sequences.
And 340, acquiring each article to be recommended, performing unique hot coding on each article to be recommended respectively, and determining an article vector matched with each article to be recommended.
And 350, respectively calculating the similarity between the user vector and each article vector, sequencing the calculation results, and determining the article corresponding to the calculation result exceeding the set threshold value as the target recommended article.
The set threshold may be 0.8, 0.9, 0.7, or the like, and is not limited in this embodiment.
In an optional implementation manner of this embodiment, after obtaining a user vector corresponding to a target user and an item vector of each item to be recommended, similarity between the user vector and each item vector may be respectively calculated, and the similarity calculation results are sorted, the item to be recommended corresponding to the similarity calculation result greater than a set threshold is determined as a target recommended item, and further, the target recommended item is recommended to the user.
According to the scheme of the embodiment, after the user vector corresponding to the target user and the item vector of each item to be recommended are obtained, the similarity between the user vector and each item vector is calculated respectively, the calculation results are sorted, the item corresponding to the calculation result exceeding the set threshold is determined as the target recommended item, and the interested item can be recommended to the user accurately.
In order to make those skilled in the art better understand the item recommendation method of the embodiment, a specific example is used for description below, and the specific process includes:
1. and extracting user behavior data.
The original behavior data of the user are related to a lot, including clicking, playing, forwarding, reading and the like, and the behaviors need to be screened according to rules, for example, reading, playing and forwarding are selected as valid data types, reading and playing with the reading time length of more than 30s or the playing time length of more than 60s are considered as valid behavior data, and forwarding data are all used as valid behaviors.
2. And sequencing the behavior data according to time and dividing the behavior data according to intervals.
And sequencing the extracted user data according to the behavior time, and meanwhile, segmenting the behaviors according to a preset interval threshold. The method specifically comprises the following steps:
1) starting from the first behavior of the user, recording the actual start _ time and the current position start _ pos of the behavior;
2) traversing subsequent behaviors, comparing the time of the behaviors with the interval of the start _ time, if the time of the behaviors is larger than a set threshold value, taking out the behaviors between the start _ pos and the current position cur _ pos, storing the behaviors into the action _ list, setting the next position as the start _ pos, and continuously traversing the behavior list;
3) after the user's behavior is traversed, resetting the start _ time and the start _ pos, and continuously traversing the behavior sequences of other users to obtain the final action _ list;
after the user behaviors are segmented, a plurality of short behavior sequences are obtained, for example, i1, i2 and i3 form one behavior sequence, i6, i7, i8 and i9 form another behavior sequence, and information and operation time intervals in all the behavior sequences are within preset values.
3. And taking all the segmented behavior sequences, and training to obtain an article vector.
The obtained article sequence contains the context information of the article, and the article can be mapped to the dense vector space by utilizing the shallow neural network to obtain an article vector. The specific method comprises the following steps:
1) carrying out one-hot encoding on the articles to obtain article vectors;
2) multiplying the item vector by a weight matrix w1 to obtain a hidden layer vector;
3) multiplying the hidden layer vector by an output weight matrix w2 to obtain an output vector;
4) comparing the output vector with the vector of the context article of the article, and updating the weight matrixes w1 and w2 according to a gradient descent method to obtain final model parameters w1 and w 2;
5) and finally multiplying the one-hot code of the article by w1 to obtain the vector of the article.
4. And (4) taking the recent behaviors (preset threshold value) of the user, and adding the item vectors to obtain a user vector.
In an optional implementation manner of this embodiment, the user vector may be obtained by using the item vector obtained in step 3. The specific method comprises the following steps:
1) presetting a threshold, for example, setting the threshold as 2 days, and extracting behavior data of the user in the last two days;
2) extracting the articles in the user behavior data, and setting weights for the articles according to specific behaviors of the user, wherein the longer the watching/reading time is, the higher the weight of the articles is;
3) and mapping the articles into article vectors, and weighting and summing the vectors according to the weight of the articles to obtain the user vectors.
5. Pushing the item to the user.
And calculating the similarity between the vector of the information to be pushed and the user vector, and preferentially pushing the articles with high similarity to the user. The method comprises the following specific steps:
1) acquiring all articles to be pushed and taking article vectors;
2) obtaining a user vector;
3) calculating cosine similarity of the user vector and the article vector;
4) and (4) sorting the articles according to the similarity, and recommending the articles with high similarity to the user.
Example four
Fig. 4 is a schematic structural diagram of an article recommendation apparatus according to a fourth embodiment of the present invention, which can execute the article recommendation methods in the embodiments. Referring to fig. 4, the apparatus includes: a user vector determination module 410, an item vector determination module 420, and a target recommended item determination module 430.
A user vector determining module 410, configured to obtain at least one effective behavior data matched with a target user, segment each effective behavior data to obtain a plurality of user behavior sequences, and determine a user vector corresponding to the target user according to each user behavior sequence;
an item vector determination module 420, configured to determine an item vector of at least one item to be recommended;
and the target recommended item determining module 430 is configured to determine a target recommended item according to the item vector of each item to be recommended and the user vector, and display the target recommended item to the user.
In the scheme of this embodiment, at least one effective behavior data matched with a target user is obtained through a user vector determination module, each effective behavior data is segmented to obtain a plurality of user behavior sequences, and a user vector corresponding to the target user is determined according to each user behavior sequence; determining an article vector of at least one article to be recommended through an article vector determination module; the target recommended article is determined by the target recommended article determining module and displayed to the user, so that the problems of low article recommending accuracy and poor article recommending effect to the user are solved, and the interested articles are accurately recommended to the user.
In an optional implementation manner of this embodiment, the user vector determination module 410 is specifically configured to
Acquiring at least one behavior data matched with the target user, and screening each behavior data according to a set threshold value to obtain each effective behavior data;
wherein the behavioral data comprises at least one of: click, play, forward, and read.
In an optional implementation manner of this embodiment, the user vector determination module 410 is further specifically configured to
Sequencing the effective behavior data according to a preset sequence, and segmenting the effective behavior data according to a preset interval threshold to obtain a plurality of user behavior sequences;
respectively inputting each user behavior sequence into a pre-trained prediction model to obtain an article vector corresponding to each user behavior sequence;
and accumulating and calculating the item vectors to generate a user vector corresponding to the target user.
In an optional implementation manner of this embodiment, the article recommending apparatus further includes: a prediction model generation module for
Acquiring training behavior sequences matched with a plurality of users;
and training a set machine learning model by using the training behavior sequence to obtain the prediction model.
In an optional implementation manner of this embodiment, the user vector determination module 410 is further specifically configured to
Recording the target starting time of the last effective behavior data of the user, and calculating the time interval between the effective starting time of the current effective behavior data and the target starting time;
and when the time interval is larger than a preset interval threshold, storing each effective behavior data in the time interval into the same behavior sequence.
In an optional implementation manner of this embodiment, the item vector determination module 420 is specifically configured to
And acquiring each article to be recommended, performing unique hot coding on each article to be recommended respectively, and determining an article vector matched with each article to be recommended.
In an optional implementation manner of this embodiment, the target recommended item determining module 430 is specifically configured to calculate similarities between the user vector and each item vector, sort calculation results, and determine an item corresponding to the calculation result exceeding a set threshold as the target recommended item.
The article recommending device provided by the embodiment of the invention can execute the article recommending method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method.
EXAMPLE five
Fig. 5 is a schematic structural diagram of a computer apparatus according to a fifth embodiment of the present invention, as shown in fig. 5, the computer apparatus includes a processor 50, a memory 51, an input device 52, and an output device 53; the number of processors 50 in the computer device may be one or more, and one processor 50 is taken as an example in fig. 5; the processor 50, the memory 51, the input device 52 and the output device 53 in the computer apparatus may be connected by a bus or other means, and the connection by the bus is exemplified in fig. 5.
The memory 51 is used as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the item recommendation method in the embodiment of the present invention (for example, the user vector determination module 410, the item vector determination module 420, and the target recommended item determination module 430 in the item recommendation device). The processor 50 executes various functional applications and data processing of the computer device by executing software programs, instructions and modules stored in the memory 51, that is, implements the above-described item recommendation method.
The memory 51 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 51 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 51 may further include memory located remotely from the processor 50, which may be connected to a computer device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 52 is operable to receive input numeric or character information and to generate key signal inputs relating to user settings and function controls of the computer apparatus. The output device 53 may include a display device such as a display screen.
EXAMPLE six
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, perform a method for item recommendation, the method including:
the method comprises the steps of obtaining at least one effective behavior data matched with a target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence;
determining an item vector of at least one item to be recommended;
and determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the method operations described above, and may also perform related operations in the item recommendation method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the article recommendation device, the included units and modules are merely divided according to the functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. An item recommendation method, comprising:
the method comprises the steps of obtaining at least one effective behavior data matched with a target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence;
determining an item vector of at least one item to be recommended;
and determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
2. The method of claim 1, wherein the obtaining at least one valid behavior data matching the target user comprises:
acquiring at least one behavior data matched with the target user, and screening each behavior data according to a set threshold value to obtain each effective behavior data;
wherein the behavioral data comprises at least one of: click, play, forward, and read.
3. The method of claim 1, wherein the segmenting each valid behavior data into a plurality of user behavior sequences comprises:
sequencing the effective behavior data according to a preset sequence, and segmenting the effective behavior data according to a preset interval threshold to obtain a plurality of user behavior sequences;
correspondingly, the determining the user vector corresponding to the target user according to each user behavior sequence includes:
respectively inputting each user behavior sequence into a pre-trained prediction model to obtain an article vector corresponding to each user behavior sequence;
and accumulating and calculating the item vectors to generate a user vector corresponding to the target user.
4. The method of claim 3, further comprising, prior to separately inputting each of the user behavior sequences into a pre-trained predictive model:
acquiring training behavior sequences matched with a plurality of users;
and training a set machine learning model by using the training behavior sequence to obtain the prediction model.
5. The method according to claim 3, wherein the segmenting each of the effective behavior data according to a preset interval threshold to obtain a plurality of user behavior sequences comprises:
recording the target starting time of the last effective behavior data of the user, and calculating the time interval between the effective starting time of the current effective behavior data and the target starting time;
and when the time interval is larger than a preset interval threshold, storing each effective behavior data in the time interval into the same behavior sequence.
6. The method of claim 1, wherein the determining an item vector for at least one item to be recommended comprises:
and acquiring each article to be recommended, performing unique hot coding on each article to be recommended respectively, and determining an article vector matched with each article to be recommended.
7. The method according to claim 1, wherein the determining a target recommended item according to the item vector of each item to be recommended and the user vector comprises:
and respectively calculating the similarity between the user vector and each article vector, sequencing the calculation results, and determining the article corresponding to the calculation result exceeding a set threshold value as the target recommended article.
8. An item recommendation device, comprising:
the user vector determining module is used for acquiring at least one effective behavior data matched with a target user, segmenting each effective behavior data to obtain a plurality of user behavior sequences, and determining a user vector corresponding to the target user according to each user behavior sequence;
the article vector determining module is used for determining an article vector of at least one article to be recommended;
and the target recommended article determining module is used for determining a target recommended article according to the article vector of each article to be recommended and the user vector, and displaying the target recommended article to the user.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the item recommendation method of any one of claims 1-7 when executing the program.
10. A storage medium containing computer-executable instructions for performing the item recommendation method of any one of claims 1-7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110351648.3A CN113065067A (en) | 2021-03-31 | 2021-03-31 | Article recommendation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110351648.3A CN113065067A (en) | 2021-03-31 | 2021-03-31 | Article recommendation method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113065067A true CN113065067A (en) | 2021-07-02 |
Family
ID=76564963
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110351648.3A Pending CN113065067A (en) | 2021-03-31 | 2021-03-31 | Article recommendation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113065067A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113590951A (en) * | 2021-07-29 | 2021-11-02 | 上海德衡数据科技有限公司 | Perception data processing method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109558514A (en) * | 2019-01-08 | 2019-04-02 | 青岛聚看云科技有限公司 | Video recommendation method, its device, information processing equipment and storage medium |
| CN110020112A (en) * | 2017-09-25 | 2019-07-16 | 北京京东尚科信息技术有限公司 | Object Push method and its system |
| CN110309427A (en) * | 2018-05-31 | 2019-10-08 | 腾讯科技（深圳）有限公司 | A kind of object recommendation method, apparatus and storage medium |
| CN110503531A (en) * | 2019-08-30 | 2019-11-26 | 中国科学技术大学 | The dynamic social activity scene recommended method of timing perception |
| CN110634047A (en) * | 2019-09-05 | 2019-12-31 | 北京无限光场科技有限公司 | Method and device for recommending house resources, electronic equipment and storage medium |
| CN111429204A (en) * | 2020-03-10 | 2020-07-17 | 携程计算机技术（上海）有限公司 | Hotel recommendation method, system, electronic equipment and storage medium |
| CN111475720A (en) * | 2020-03-31 | 2020-07-31 | 北京三快在线科技有限公司 | Recommendation method, recommendation device, server and storage medium |
| CN111797319A (en) * | 2020-07-01 | 2020-10-20 | 喜大（上海）网络科技有限公司 | Recommendation method, device, equipment and storage medium |
| CN112001776A (en) * | 2020-08-24 | 2020-11-27 | 上海风秩科技有限公司 | Service information pushing method and device, storage medium and electronic equipment |
-
2021
- 2021-03-31 CN CN202110351648.3A patent/CN113065067A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020112A (en) * | 2017-09-25 | 2019-07-16 | 北京京东尚科信息技术有限公司 | Object Push method and its system |
| CN110309427A (en) * | 2018-05-31 | 2019-10-08 | 腾讯科技（深圳）有限公司 | A kind of object recommendation method, apparatus and storage medium |
| CN109558514A (en) * | 2019-01-08 | 2019-04-02 | 青岛聚看云科技有限公司 | Video recommendation method, its device, information processing equipment and storage medium |
| CN110503531A (en) * | 2019-08-30 | 2019-11-26 | 中国科学技术大学 | The dynamic social activity scene recommended method of timing perception |
| CN110634047A (en) * | 2019-09-05 | 2019-12-31 | 北京无限光场科技有限公司 | Method and device for recommending house resources, electronic equipment and storage medium |
| CN111429204A (en) * | 2020-03-10 | 2020-07-17 | 携程计算机技术（上海）有限公司 | Hotel recommendation method, system, electronic equipment and storage medium |
| CN111475720A (en) * | 2020-03-31 | 2020-07-31 | 北京三快在线科技有限公司 | Recommendation method, recommendation device, server and storage medium |
| CN111797319A (en) * | 2020-07-01 | 2020-10-20 | 喜大（上海）网络科技有限公司 | Recommendation method, device, equipment and storage medium |
| CN112001776A (en) * | 2020-08-24 | 2020-11-27 | 上海风秩科技有限公司 | Service information pushing method and device, storage medium and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 吴梅梅: "机器学习算法及其应用", 31 May 2020, 机械工业出版社, pages: 57 - 58 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113590951A (en) * | 2021-07-29 | 2021-11-02 | 上海德衡数据科技有限公司 | Perception data processing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651542B (en) | Article recommendation method and device | |
| CN111339415B (en) | Click rate prediction method and device based on multi-interactive attention network | |
| CN109831684B (en) | Video optimization recommendation method and device and readable storage medium | |
| CN110163647B (en) | Data processing method and device | |
| CN108205766A (en) | Information-pushing method, apparatus and system | |
| CN108021708B (en) | Content recommendation method and device and computer readable storage medium | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN111444334A (en) | Data processing method, text recognition device and computer equipment | |
| US20230066853A1 (en) | Method and apparatus for training information prediction models, method and apparatus for predicting information, and storage medium and device thereof | |
| CN108305181B (en) | Social influence determination method and device, information delivery method and device, equipment and storage medium | |
| CN111695960A (en) | Object recommendation system, method, electronic device and storage medium | |
| CN110543603B (en) | Collaborative filtering recommendation method, device, equipment and medium based on user behaviors | |
| CN111967914A (en) | User portrait based recommendation method and device, computer equipment and storage medium | |
| CN113268656A (en) | User recommendation method and device, electronic equipment and computer storage medium | |
| CN111400586A (en) | Group display method, terminal, server, system and storage medium | |
| CN113837842A (en) | Commodity recommendation method and equipment based on user behavior data | |
| CN114896517A (en) | Commodity recommendation method, system, equipment and storage medium | |
| CN112100221A (en) | Information recommendation method and device, recommendation server and storage medium | |
| CN115147130A (en) | Problem prediction method, apparatus, storage medium, and program product | |
| WO2021081914A1 (en) | Pushing object determination method and apparatus, terminal device and storage medium | |
| CN111523315B (en) | Data processing method, text recognition device and computer equipment | |
| CN115935185A (en) | Training method and device for recommendation model | |
| CN117391824B (en) | Method and device for recommending articles based on large language model and search engine | |
| CN112989182B (en) | Information processing method, information processing device, information processing apparatus, and storage medium | |
| CN111680213A (en) | Information recommendation method, data processing method and device |
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 |