CN108363730B - Content recommendation method, system and terminal equipment - Google Patents
Content recommendation method, system and terminal equipment Download PDFInfo
- Publication number
- CN108363730B CN108363730B CN201810032233.8A CN201810032233A CN108363730B CN 108363730 B CN108363730 B CN 108363730B CN 201810032233 A CN201810032233 A CN 201810032233A CN 108363730 B CN108363730 B CN 108363730B
- Authority
- CN
- China
- Prior art keywords
- preset
- time
- label
- sharing
- program
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Abstract
The invention is suitable for the technical field of information processing, and provides a content recommendation method, a content recommendation system and terminal equipment, wherein the content recommendation method comprises the following steps: acquiring a historical watching record of a user; obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record; obtaining an adding factor and an attenuation factor corresponding to a preset label according to the historical watching records; acquiring a program recommendation list corresponding to a preset label, and assigning programs in the program recommendation list to obtain recommendation assignment; and obtaining the recommended content according to the time-sharing and grading label weight, the adding and grading factor, the attenuation factor and the recommendation assignment. According to the embodiment of the invention, the interesting degree of the user to the preset label in the preset time period is calculated by considering the program which the user is interested in the preset time period, and the recommended content pushed to the user in the preset time period is calculated, so that the specific recommended content is pushed to the user more accurately, and the user experience is optimized.
Description
Technical Field
The invention belongs to the technical field of information processing, and particularly relates to a content recommendation method, a content recommendation system and terminal equipment.
Background
With the development of internet technology, users watch movie programs by applying more intelligent terminals, and generally, a method for searching programs by a user includes searching favorite programs of the user through recommended contents or directly searching known programs for watching.
At present, methods for searching programs liked by a user through recommended content include a collaborative filtering recommendation algorithm based on content and a collaborative filtering recommendation algorithm based on user behavior, wherein recommendation results based on the collaborative filtering recommendation algorithm are fixed, and recommendation results are easy to change; the recommendation result of the collaborative filtering recommendation algorithm based on the user behavior is recommended according to the user programs close to the user, the recommendation result can be updated frequently, but the calculation amount is large.
Disclosure of Invention
In view of this, embodiments of the present invention provide a content recommendation method, system and terminal device, so as to solve the problems in the prior art that a recommendation result is unchanged and a calculation amount is large.
A first aspect of an embodiment of the present invention provides a content recommendation method, including:
acquiring a historical watching record of a user;
obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record;
obtaining an adding factor and an attenuation factor corresponding to a preset label according to the historical watching records;
acquiring a program recommendation list corresponding to a preset label, and assigning programs in the program recommendation list to obtain recommendation assignment;
and obtaining the recommended content according to the time-sharing and grading label weight, the adding and grading factor, the attenuation factor and the recommendation assignment.
A second aspect of an embodiment of the present invention provides a content recommendation system, including:
the historical watching record acquisition module is used for acquiring the historical watching record of the user;
the time-sharing label weight acquisition module is used for acquiring a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record;
the influence factor acquisition module is used for acquiring an adding factor and an attenuation factor corresponding to a preset label according to the historical watching record;
the recommended program assignment module is used for acquiring a program recommendation list corresponding to the preset label and assigning the programs in the program recommendation list to obtain a recommendation assignment;
and the recommended content acquisition module is used for obtaining recommended content according to the time-sharing tagging weight, the adding factor, the attenuation factor and the recommended assignment.
A third aspect of the embodiments of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the content recommendation method as described above when executing the computer program.
A fourth aspect of embodiments of the present invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps of the content recommendation method described above.
Compared with the prior art, the embodiment of the invention has the following beneficial effects: obtaining a historical watching record of a user; obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record; obtaining an adding factor and an attenuation factor corresponding to a preset label according to the historical watching records; acquiring a program recommendation list corresponding to a preset label, and assigning programs in the program recommendation list to obtain recommendation assignment; and obtaining the recommended content according to the time-sharing and grading label weight, the adding and grading factor, the attenuation factor and the recommendation assignment. According to the embodiment of the invention, the interesting degree of the user to the preset label in the preset time period is calculated by considering the program which the user is interested in the preset time period, and the recommended content pushed to the user in the preset time period is calculated, so that the specific recommended content is pushed to the user more accurately, and the user experience is optimized.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a schematic flow chart of an implementation of a content recommendation method provided in an embodiment of the present invention;
fig. 2 is a schematic flow chart of an implementation of step S102 in fig. 1 according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of an implementation of step S103 in fig. 1 according to an embodiment of the present invention;
fig. 4 is a schematic flow chart of the implementation of step S303 in fig. 3 according to an embodiment of the present invention;
fig. 5 is a schematic flow chart of an implementation of step S306 in fig. 3 according to an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of a content recommendation system according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a time division label weight obtaining module in fig. 6 according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an impact factor obtaining module in fig. 6 according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of the full-time impact factor obtaining unit in fig. 8 according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of the time-sharing impact factor obtaining unit in fig. 8 according to an embodiment of the present invention;
fig. 11 is a schematic diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
The terms "comprises" and "comprising," and any variations thereof, in the description and claims of this invention and the above-described drawings are intended to cover non-exclusive inclusions. For example, a process, method, or system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus. Furthermore, the terms "first," "second," and "third," etc. are used to distinguish between different objects and are not used to describe a particular order.
In order to explain the technical means of the present invention, the following description will be given by way of specific examples.
Example 1:
fig. 1 shows an implementation flow of a content recommendation method according to an embodiment of the present invention, where a flow execution subject of this embodiment may be a terminal device, and a process thereof is detailed as follows:
in step S101, a history viewing record of the user is acquired.
In this embodiment, a history viewing record of a preset time duration is obtained, where the history viewing record includes a plurality of programs arranged in chronological order.
In step S102, a preset time period and a time-sharing label weight corresponding to a preset label are obtained according to the historical viewing record.
In this embodiment, the preset time period may be a morning time period (00: 00-6: 00), a morning time period (6: 00-11: 00), a midday time period (11: 00-13: 00), an afternoon time period (13: 00-17: 00), an evening time period (17: 00-19: 00), a night golden time period (19: 00-22: 00), and a night time period (22: 00-24: 00).
In this embodiment, each program in the history record carries at least one preset tag. The preset labels may include antiques, cities, fantasy, love, war, action, and the like.
In step S103, a scoring factor and an attenuation factor corresponding to the preset tag are obtained according to the historical viewing record.
In step S104, a program recommendation list corresponding to the preset tag is obtained, and the programs in the program recommendation list are assigned to obtain recommendation assignments.
In step S105, the recommended content is obtained according to the time-sharing tagging weight, the adding factor, the attenuating factor, and the recommendation assignment.
According to the embodiment, the historical watching records of the user are obtained; obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record; obtaining an adding factor and an attenuation factor corresponding to a preset label according to the historical watching records; acquiring a program recommendation list corresponding to a preset label, and assigning programs in the program recommendation list to obtain recommendation assignment; and obtaining the recommended content according to the time-sharing and grading label weight, the adding and grading factor, the attenuation factor and the recommendation assignment. According to the embodiment of the invention, the interesting degree of the user to the preset label in the preset time period is calculated by considering the program which the user is interested in the preset time period, and the recommended content pushed to the user in the preset time period is calculated, so that the specific recommended content is pushed to the user more accurately, and the user experience is optimized.
As shown in fig. 2, in an embodiment of the present invention, fig. 2 shows a specific implementation flow of step S102 in fig. 1, which is detailed as follows:
in step S201, a program corresponding to a preset time period is extracted from the historical viewing record, so as to obtain a time-sharing viewing sequence.
In this embodiment, the programs in the history viewing record include viewing time, and the time-sharing viewing sequence is extracted according to the viewing time and a preset time period. According to the preset time period, the historical watching records can be divided into a morning time-sharing watching sequence, a noon time-sharing watching sequence, an afternoon time-sharing watching sequence, an evening time-sharing watching sequence, a night gold time-sharing watching sequence and a night time-sleeping time-sharing watching sequence.
In step S202, a program corresponding to a preset tag is extracted from the time-sharing viewing sequence, so as to obtain a time-sharing tag viewing sequence.
In this embodiment, when the time-sharing viewing sequence is obtained, a time-sharing tag viewing sequence is obtained according to a preset tag, and each tag corresponds to one time-sharing tag viewing sequence.
Taking a specific application scenario as an example:
in the historical watching records, the preset label of the program 'mei highway action' is action and crime, and the watching time is 8: 00; the preset labels of the program 'Yangli Jiangshan long singing line' are love and ancient clothes, the watching time is 14:00, the preset labels of Sansheng Sanshijiri peach flowers are ancient clothes, fantasy and love, the watching time is 15:00, the preset labels of warwolf 2 are wars and actions, and the watching time is 16: 30.
The time-sharing viewing sequence obtained by taking the preset time interval as afternoon is the long singing line of the beautiful river mountain, the peach blossom in the third generation and the tenth of the third generation and the warwolf 2.
According to the time-sharing watching sequence, a preset label is taken as a love, and the time-sharing and mark-sharing watching sequence is as follows: the long singing line of beautiful Jiangshan and the ten-miles of the three generations of peach blossom.
In step S203, the total number of programs in the time-sharing tagging viewing sequence is divided by the total number of programs in the time-sharing viewing sequence to obtain the weight of the time-sharing tagging.
In this embodiment, if the preset tag is i and the preset time period is t, the total number of programs in the view sequence of the time-sharing tag is Num (n)t,i) The total number of programs in the time-sharing viewing sequence is Num (n)t) Then time-division label weight Wt,iIs composed of
Wt,iAnd representing the time-sharing label weight corresponding to the preset label i in the preset time period t.
As shown in fig. 3, in an embodiment of the present invention, fig. 3 shows an implementation flow of step S103 in fig. 1, which is detailed as follows;
in step S301, a value is assigned to a program in the historical viewing record according to a first preset rule, so as to obtain a program score corresponding to each program.
In this embodiment, the assigning values to the programs in the historical viewing record according to the first preset rule includes assigning values to the programs in the historical viewing record incrementally with a preset tolerance according to a chronological order. For example, the score of the program ranked at the first position of the historical viewing record is 1, and the score of the program ranked at the second position is 2, so that the programs in the historical viewing record are sequentially and incrementally assigned with 1 as a preset tolerance.
In step S302, a program corresponding to a preset tag is extracted from the historical viewing record, so as to obtain a tag viewing sequence.
In step S303, a full-time adding factor and a full-time attenuating factor corresponding to the preset tag are calculated according to the program score and the view sequence of the sub-tag.
In step S304, a program corresponding to a preset time period is extracted from the historical viewing record, so as to obtain a time-sharing viewing sequence.
In step S305, a program corresponding to the preset tag is extracted from the time-sharing viewing sequence, so as to obtain a time-sharing tag viewing sequence.
In step S306, a time-sharing plus-scoring factor and a time-sharing decay factor corresponding to the preset tag are calculated according to the program score, the time-sharing viewing sequence, and the time-sharing tag viewing sequence.
As shown in fig. 4, in an embodiment of the present invention, fig. 4 shows an implementation flow of step S303 in fig. 3, which is detailed as follows:
in step S401, the sum of the program scores corresponding to the programs in the tab viewing sequence is calculated to obtain the sum of the tab program scores.
In step S402, based on a first preset calculation rule, a full-time attenuation factor corresponding to a preset tag is calculated according to the score sum of the sub-tag programs, the total number of the programs in the viewing sequence of the sub-tag, the total number of the programs in the historical viewing record, and a first preset attenuation coefficient.
In this embodiment, the first preset calculation rule is:
ai＝1+d1*(nL*Num(ni)-Sum(ni))
wherein, aiSetting a full-time attenuation factor corresponding to a preset label i; d1A first predetermined attenuation factor; n isLFor the total number of programs in the history viewing record, n is the total number of programs in the present embodimentLThe value can be 10, which is used for representing the programs that the user likes to watch in the near period of time; num (n)i) The total number of programs in the view sequence is received for the label; sum (n)i) Is the sum of the scores of the sub-label programs.
And obtaining a full-time attenuation factor corresponding to the preset label through the first preset calculation rule.
In step S403, when the preset labels corresponding to N consecutive programs in the historical viewing record are the same, obtaining a value N, where N is greater than or equal to 2.
In this embodiment, when the programs at the N consecutive positions in the history viewing record include the same preset tag, it indicates that there is a continuous viewing at this time, and the user is more interested in the preset tag, so that the N value is obtained, and the full-time bonus factor corresponding to the preset tag is calculated.
In step S404, the N value is multiplied by a first preset bonus coefficient to obtain a full-time bonus factor corresponding to the preset tag.
In this embodiment, the full time bonus factorThe calculation formula of (2) is as follows: bi＝a1N, wherein a1Is a first preset bonus coefficient; biAnd adding a score factor for the preset label i.
As shown in fig. 5, in an embodiment of the present invention, fig. 5 is an implementation flow of step S306 in fig. 3, which is detailed as follows:
in step S501, the sum of the program scores corresponding to the programs in the view-receiving sequence of the time-sharing tag is calculated, so as to obtain the sum of the program scores of the time-sharing tag.
In step S502, based on a second preset calculation rule, a time-sharing attenuation factor corresponding to the preset tag is calculated according to the time-sharing tag program score sum, the total number of programs in the time-sharing tag viewing sequence, the total number of programs in the historical viewing record, and a second preset attenuation coefficient.
In this embodiment, the second preset calculation rule is:
at,i＝1+d2*(nL*Num(nt,i)-Sum(nt,i))；
wherein d is2A second predetermined attenuation factor; num (n)t,i) The total number of programs in the view sequence is received for the time-sharing label-dividing; sum (n)t,i) Is the sum of the time-sharing label program scores. Obtaining the time-sharing attenuation factor a through the second preset calculation rulet,i
In step S503, when the preset tags corresponding to M consecutive programs in the time-sharing viewing sequence are the same, a value M is obtained, where M is greater than or equal to 2.
In this embodiment, when the programs at the M consecutive positions in the time-sharing viewing sequence include the same preset tag, it indicates that there is a continuous viewing in the preset time period, and the user is more interested in the preset tag in the preset time period, so that the value M is obtained, and the time-sharing bonus factor of the preset tag is calculated.
In step S504, the value M is multiplied by a second preset bonus coefficient to obtain a time-sharing bonus factor corresponding to the preset tag.
In this embodiment, step S504 includes: bt,i＝a2M, wherein a2Is a second predetermined addendum coefficient, bt,iAnd presetting a time-sharing bonus factor corresponding to the label i.
In this embodiment, the specific process of step S104 in fig. 1 includes:
in step S601, a common program recommendation list is acquired.
In this embodiment, the server pushes a public program recommendation list to the terminal device, the server includes an internal recommendation system, and the preset algorithm of the internal recommendation system includes a content-based collaborative filtering recommendation algorithm, a user behavior-based collaborative filtering recommendation algorithm, a hit play recommendation algorithm, and a hit search recommendation algorithm.
In this embodiment, the public program recommendation list further includes a program list acquired based on an external data source, such as a bean program ranking list, a popularity movie audience rating index recommendation list, and the like.
In this embodiment, the obtained public program recommendation lists are different according to different preset algorithms, and if the preset algorithm is multiple, the obtained public program recommendation lists are multiple.
Step S602: and extracting all programs corresponding to the preset labels from the public program recommendation list to obtain a program recommendation list corresponding to each preset label.
In this embodiment, a program recommendation list corresponding to each preset tag is obtained by extracting all programs corresponding to the preset tags.
Step S603: and assigning the programs in the program recommendation list according to a second preset rule to obtain recommendation assignment corresponding to each program in the public program recommendation list.
In this embodiment, step S603 includes: and arranging the programs in the program recommendation list corresponding to each preset label according to the arrangement sequence of the programs in the public program recommendation list, and assigning the programs in the program recommendation list in a descending manner according to preset tolerance. For example, with 1 as a preset tolerance, the program value of the first position in the program recommendation list corresponding to all the preset labels is assigned to 100, the program value of the second position is assigned to 99, and so on, so as to obtain the recommendation assignment of all the programs in the common program recommendation list.
In an embodiment of the present invention, a specific implementation flow of step S105 in fig. 1 includes:
step S701: and performing integral operation on the branch label weight, the scoring factor, the attenuation factor and the recommendation assignment based on a third preset calculation rule to obtain a recommendation score of the program in the public program recommendation list.
In an embodiment of the present invention, the content recommendation method further includes:
and acquiring the occurrence frequency of each program in all the public recommendation lists and the number of the preset labels.
In this embodiment, since there are a plurality of preset algorithms, there are also a plurality of corresponding public program recommendation lists, and the number of times of occurrence of each program in all the public recommendation lists is obtained.
In this embodiment, taking a program p in a public program recommendation list as an example, calculating a recommendation score of the program p specifically includes:
the third preset calculation rule is as follows:
wherein n ispRepresenting the number of occurrences of program p in all public recommendation lists; n isx,pThe number of preset tags representing program p; i represents a preset label i; t represents a preset time period t; a ist,iRepresenting a time-sharing attenuation factor corresponding to a preset label i; a isiRepresenting a full-time attenuation factor corresponding to a preset label i; bt,iRepresenting a time-sharing and scoring factor corresponding to a preset label i; biRepresenting a full time adding factor corresponding to a preset label i; c represents a preset bonus item; w is at,iRepresenting time-division tagging weights; dx,i,pAnd representing the recommendation assignment of the program p obtained by screening by adopting a preset algorithm x in the program recommendation list corresponding to the preset label i.
In this embodiment, the preset bonus item c includes a first preset bonus value and a second preset bonus value, and the obtaining process of the numerical value of the preset bonus item c includes:
1) and c is a first preset added value when one preset label corresponding to the program p is the same as one preset label corresponding to the program at the first position in the historical viewing record.
2) And c is a second preset added value when the preset label corresponding to the program p is different from all the preset labels corresponding to the programs at the first position in the historical viewing record.
In this embodiment, the first preset bonus value may be set to 0.02 and the second preset bonus value may be set to 0. By acquiring the score adding item value, the preset label which is currently interested by the user can be added, so that the recommended content can more cater to the current interest of the user.
And calculating to obtain recommendation scores of all programs in the public program recommendation list according to the third preset calculation rule.
Step S702: and correspondingly arranging the programs in the public program recommendation list according to the recommendation scores from high to low to obtain recommended contents.
In this embodiment, the programs in the common program recommendation list are sorted in the order from high recommendation score to low recommendation score, and the programs with the preset recommendation number are obtained from the program with the highest recommendation score to the next program as the recommendation content.
In an embodiment of the present invention, when the programs in the acquired recommended content include programs in the historical viewing record, the programs that are the same as the programs in the historical viewing record are removed, and the program with the highest recommendation score is selected from the programs remaining in the public recommendation list and placed in the recommended content, and the like is performed until the programs in the historical viewing record of the user do not exist in the recommended content. Thereby ensuring that the recommended content is content that is not seen and may be of interest to the user.
It can be known from the above embodiments that by obtaining the tag weighting, the scoring factor and the attenuation factor, the predetermined tags continuously watched by the user are scored, and the tags currently interested by the user are scored, so that the recommended content recommended to the user more caters to the current interest of the user, and according to the predetermined time period, the historical viewing records of the user in different predetermined time periods are divided, so that the recommended content interested by the user in different time periods can be obtained, and the user is prevented from missing the interested content.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
Example 2:
as shown in fig. 6, a content recommendation system 1000 according to an embodiment of the present invention is configured to perform the method steps in the embodiment corresponding to fig. 1, and includes:
a historical viewing record obtaining module 1100, configured to obtain a historical viewing record of a user;
a time-division label weight obtaining module 1200, configured to obtain a preset time period and a time-division label weight corresponding to a preset label according to the historical viewing record;
the influence factor obtaining module 1300 is configured to obtain an adding factor and an attenuation factor corresponding to a preset tag according to the historical viewing record;
the recommended program assignment module 1400 is configured to obtain a program recommendation list corresponding to a preset tag, and assign a value to a program in the program recommendation list to obtain a recommendation assignment;
and a recommended content obtaining module 1500, configured to obtain recommended content according to the time-sharing tagging weight, the adding factor, the attenuating factor, and the recommended assignment.
According to the embodiment, the historical watching records of the user are obtained; obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record; obtaining an adding factor and an attenuation factor corresponding to a preset label according to the historical watching records; acquiring a program recommendation list corresponding to a preset label, and assigning programs in the program recommendation list to obtain recommendation assignment; and obtaining the recommended content according to the time-sharing and grading label weight, the adding and grading factor, the attenuation factor and the recommendation assignment. According to the embodiment of the invention, the interesting degree of the user to the preset label in the preset time period is calculated by considering the program which the user is interested in the preset time period, and the recommended content pushed to the user in the preset time period is calculated, so that the specific recommended content is pushed to the user more accurately, and the user experience is optimized.
As shown in fig. 7, in an embodiment of the present invention, the time-division label weight obtaining module 1200 in the embodiment corresponding to fig. 6 shown in fig. 7 further includes a structure for executing the method steps in the embodiment corresponding to fig. 2, which includes:
a time-sharing viewing sequence obtaining unit 1210, configured to extract the program corresponding to the preset time period from the historical viewing record, so as to obtain a time-sharing viewing sequence;
a time-division label viewing sequence obtaining unit 1220, configured to extract a program corresponding to the preset label from the time-division viewing sequence, so as to obtain a time-division label viewing sequence;
the time-sharing tag weight obtaining unit 1230 is configured to divide the total number of programs in the time-sharing tag viewing sequence by the total number of programs in the time-sharing viewing sequence to obtain a time-sharing tag weight.
As shown in fig. 8, in an embodiment of the present invention, the impact factor obtaining module 1300 in the embodiment corresponding to fig. 6 and output in fig. 8 is further configured to execute the structure of the method step in the embodiment corresponding to fig. 3, where the structure includes:
a program score obtaining unit 1310, configured to assign a value to a program in the historical viewing record according to a first preset rule, so as to obtain a program score corresponding to each program;
a tag viewing sequence acquiring unit 1320, configured to extract a program corresponding to a preset tag from a historical viewing record, so as to obtain a tag viewing sequence;
a full-time influence factor obtaining unit 1330, configured to calculate a full-time adding and dividing factor and a full-time attenuating factor corresponding to the preset tag according to the program score and the view sequence of the sub-tag;
the time-sharing viewing sequence obtaining unit 1340 is configured to extract a program corresponding to a preset time period from the historical viewing record, so as to obtain a time-sharing viewing sequence;
a time-division label viewing sequence acquiring unit 1350 configured to extract a program corresponding to a preset label from the time-division viewing sequence to obtain a time-division label viewing sequence;
the time-sharing influence factor acquiring unit 1360 is configured to calculate a time-sharing plus-scoring factor and a time-sharing decay factor corresponding to the preset tag according to the program score, the time-sharing viewing sequence, and the time-sharing tag viewing sequence.
As shown in fig. 9, in an embodiment of the present invention, the full-time impact factor obtaining unit 1330 in fig. 8 shown in fig. 9 is further configured to execute the method steps in the embodiment corresponding to fig. 4, including:
a sub-label program score sum calculation subunit 1331, configured to calculate a sum of program scores corresponding to programs in the sub-label viewing sequence, to obtain a sub-label program score sum;
the full-time attenuation factor calculating subunit 1332 is configured to calculate, based on a first preset calculation rule, a full-time attenuation factor corresponding to the preset tag according to the sum of the scores of the sub-tag programs, the total number of the programs in the sub-tag viewing sequence, the total number of the programs in the historical viewing record, and a first preset attenuation coefficient;
an N value obtaining subunit 1333, configured to obtain the N value when preset tags corresponding to N consecutive programs are the same in the history viewing record, where N is greater than or equal to 2;
and a full-time bonus factor obtaining subunit 1334, configured to multiply the N value by the first preset bonus coefficient, so as to obtain a full-time bonus factor corresponding to the preset tag.
As shown in fig. 10, in an embodiment of the present invention, the structure of the time-sharing influence factor acquiring unit 1360 shown in fig. 8 and shown in fig. 10 is further configured to perform the method steps in the embodiment corresponding to fig. 5, including:
a time-division label program score sum obtaining subunit 1361, configured to calculate a sum of program scores corresponding to programs in the time-division label viewing sequence to obtain a time-division label program score sum;
the time-sharing attenuation factor calculation subunit 1362 is configured to calculate, based on a second preset calculation rule, to obtain a time-sharing attenuation factor corresponding to the preset tag according to the time-sharing tag program value sum, the total number of programs in the viewing sequence of the time-sharing tag, the total number of programs in the historical viewing record, and a second preset attenuation factor;
an M value obtaining subunit 1363, configured to obtain, when preset tags corresponding to M consecutive programs in the time-sharing viewing sequence are the same, the M value, where M is greater than or equal to 2;
and a time-sharing bonus factor acquiring subunit 1364, configured to multiply the M value by the second preset bonus coefficient to obtain a time-sharing bonus factor corresponding to the preset tag.
In one embodiment of the present invention, the module 1400 in fig. 6 for assigning recommended programs includes:
a public program recommendation list obtaining unit, configured to obtain a public program recommendation list;
the program recommendation list acquisition unit is used for extracting all programs corresponding to the preset labels from the public program recommendation list to obtain a program recommendation list corresponding to each preset label;
and the recommendation assignment obtaining unit is used for assigning the programs in the program recommendation list according to a second preset rule to obtain a recommendation assignment corresponding to each program in the public program recommendation list.
In an embodiment of the present invention, the recommended content obtaining module 1500 in fig. 6 further includes:
the recommendation score calculating unit is used for performing integral operation on the number of preset algorithms, the sub-label weights, the adding and dividing factors, the attenuation factors and the recommendation assignment values based on a third preset calculating rule to obtain recommendation scores of programs in the public program recommendation list;
and the recommended content acquisition unit is used for correspondingly arranging the programs in the public program recommendation list according to the recommendation scores from high to low to obtain recommended content.
It can be known from the above embodiments that by obtaining the tag weighting, the scoring factor and the attenuation factor, the predetermined tags continuously watched by the user are scored, and the tags currently interested by the user are scored, so that the recommended content recommended to the user more caters to the current interest of the user, and according to the predetermined time period, the historical viewing records of the user in different predetermined time periods are divided, so that the recommended content interested by the user in different time periods can be obtained, and the user is prevented from missing the interested content.
Example 3:
the embodiment of the present invention further provides a terminal device 11, which includes a memory 111, a processor 110, and a computer program 112 stored in the memory 111 and operable on the processor 110, where the processor 110 executes the computer program 112 to implement steps in each embodiment described in embodiment 1, for example, steps S101 to S105 shown in fig. 1. Alternatively, the processor 110, when executing the computer program 112, implements the functions of the modules in the device embodiments as described in embodiment 2, for example, the functions of the modules 1100 to 1500 shown in fig. 6.
The terminal device 11 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The terminal device 11 may include, but is not limited to, a processor, a memory. For example, the terminal device 11 may further include an input/output device, a network access device, a bus, and the like.
The Processor 110 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 111 may be an internal storage unit of the terminal device 11, such as a hard disk or a memory of the terminal device 11. The memory 111 may also be an external storage device of the terminal device 11, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the terminal device 11. Further, the memory 11 may also include both an internal storage unit of the terminal device 11 and an external storage device. The memory 111 is used for storing the computer program 112 and other programs and data required by the terminal device 11. The memory 111 may also be used to temporarily store data that has been output or is to be output.
Example 4:
an embodiment of the present invention further provides a computer-readable storage medium, in which a computer program 112 is stored, and when being executed by a processor, the computer program implements the steps in the embodiments described in embodiment 1, for example, step S101 to step S105 shown in fig. 1. Alternatively, the computer program realizes the functions of the respective modules in the respective apparatus embodiments as described in embodiment 2, for example, the functions of the modules 1100 to 1500 shown in fig. 6, when being executed by a processor.
The computer program 112 may be stored in a computer readable storage medium, and when executed by the processor 110, the computer program 112 may implement the steps of the above-described method embodiments. Wherein the computer program 112 comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, etc. It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media which may not include electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
The steps in the method of the embodiment of the invention can be sequentially adjusted, combined and deleted according to actual needs.
The modules or units in the system of the embodiment of the invention can be combined, divided and deleted according to actual needs.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (6)
1. A content recommendation method, comprising:
acquiring a historical watching record of a user;
obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical watching record, wherein the preset label represents the program category;
obtaining an adding factor and an attenuation factor corresponding to the preset label according to the historical watching record, wherein the adding factor comprises a full-time adding factor and a time-sharing adding factor, and the attenuation factor comprises a full-time attenuation factor and a time-sharing attenuation factor;
acquiring a program recommendation list corresponding to the preset label, and assigning the programs in the program recommendation list to obtain recommendation assignment;
obtaining recommended content according to the time-sharing tagging weight, the adding factor, the attenuation factor and the recommendation assignment;
the obtaining of the bonus factor and the attenuation factor corresponding to the preset tag according to the historical viewing record includes:
assigning values to the programs in the historical viewing records according to a first preset rule to obtain a program score corresponding to each program, wherein the assigning values to the programs in the historical viewing records according to the first preset rule comprise assigning values to the programs in the historical viewing records in an increasing manner according to a time sequence and preset tolerance;
extracting programs corresponding to the preset tags from the historical viewing records to obtain a sub-tag viewing sequence;
calculating to obtain the full-time adding and dividing factor and the full-time attenuation factor corresponding to the preset label according to the program score and the view sequence of the sub-label;
extracting the programs corresponding to the preset time period from the historical viewing records to obtain a time-sharing viewing sequence;
extracting the program corresponding to the preset label from the time-sharing viewing sequence to obtain a time-sharing viewing sequence of the time-sharing viewing label;
calculating to obtain the time-sharing adding and dividing factor and the time-sharing attenuation factor corresponding to the preset label according to the program score, the time-sharing viewing sequence and the time-sharing label viewing sequence;
the step of calculating the full-time point-adding factor of the preset label and the full-time attenuation factor of the preset label according to the program score and the view sequence of the sub-label comprises the following steps:
calculating the sum of the program scores corresponding to the programs in the sub-label viewing sequence to obtain the sum of the sub-label program scores;
based on a first preset calculation rule, calculating to obtain a full-time attenuation factor corresponding to the preset label according to the score sum of the sub-label programs, the total number of the programs in the sub-label viewing sequence, the total number of the programs in the historical viewing record and a first preset attenuation coefficient;
when the preset labels corresponding to N continuous programs in the historical watching records are the same, obtaining the value N, wherein N is more than or equal to 2;
multiplying the N value by a first preset adding coefficient to obtain a full-time adding factor corresponding to the preset label;
the first preset calculation rule is as follows:
ai＝1+d1*(nL*Num(ni)-Sum(ni))
wherein, aiFor a full time attenuation factor, d, corresponding to a preset label i1Is a first predetermined attenuation system, nLFor the total number of programs in the historical viewing record, Num (n)i) For the total number of programs in the tab view, Sum (n)i) Is the sum of the scores of the sub-label programs;
the calculation formula of the full-time addend factor is as follows:
bi＝a1*N
wherein, a1Is a first predetermined addendum coefficient, biAdding a score factor for the preset label i;
the step of calculating the time-sharing bonus factor and the time-sharing attenuation factor according to the program score, the time-sharing viewing sequence and the time-sharing tag viewing sequence comprises the following steps:
calculating the sum of the program scores corresponding to the programs in the time-sharing tag watching sequence to obtain the sum of the time-sharing tag program scores;
calculating to obtain a time-sharing attenuation factor corresponding to the preset label according to the time-sharing label program score sum, the total number of programs in the time-sharing label viewing sequence, the total number of programs in the historical viewing record and a second preset attenuation coefficient on the basis of a second preset calculation rule;
when the preset labels corresponding to M continuous programs in the time-sharing viewing sequence are the same, obtaining the value M, wherein M is more than or equal to 2;
multiplying the M value by the second preset bonus coefficient to obtain a time-sharing bonus factor corresponding to the preset label;
the second preset calculation rule is as follows:
at，i＝1+d2*(nL*Num(nt，i)-Sum(nt，i))
wherein, at，iTime-sharing attenuation factor, d2For the second predetermined attenuation coefficient, Num (n)t，i) For the total number of programs in the time-division label-receiving view sequence, Sum (n)t，i) Is the sum of the time-sharing label program scores;
the calculation formula of the time-sharing bonus factor corresponding to the preset label is as follows:
bt，i＝a2*M
wherein, bt，iIs a time-sharing bonus factor, a, corresponding to a preset label i2A second preset bonus coefficient;
obtaining recommendation content according to the time-sharing tagging weight, the adding factor, the attenuation factor and the recommendation assignment, wherein the obtaining of the recommendation content comprises:
based on a third preset calculation rule, calculating to obtain a recommendation score of the programs in the public program recommendation list according to the time-sharing sub-label weight, the adding sub-factor, the attenuation factor and the recommendation assignment;
correspondingly arranging the programs in the public program recommendation list according to the recommendation score from high to low to obtain recommended content;
the third preset calculation rule is as follows:
wherein n ispIndicating the number of occurrences of program p, n, in all public recommendation listsx，pNumber of preset tags representing program p, i represents preset tag i, t represents preset time period t, at，iRepresents a time-sharing attenuation factor corresponding to a preset label i, aiRepresents the full-time attenuation factor corresponding to the preset label i, bt，iRepresenting the time-sharing and scoring factor corresponding to the preset label i, biRepresenting the full time bonus factor corresponding to the preset label i, c representing the preset bonus item, wt，iRepresenting time-of-use sub-label weights, Dx，i，pRepresenting recommendation assignment of the program p obtained by screening by adopting a preset algorithm x in a program recommendation list corresponding to a preset label i;
the preset adding item c comprises a first preset adding value and a second preset adding value, and the obtaining process of the numerical value of the preset adding item c comprises the following steps:
when one preset label corresponding to the program p is the same as one preset label corresponding to the program at the first position in the historical viewing record, c is a first preset added value;
and c is a second preset added value when the preset label corresponding to the program p is different from all the preset labels corresponding to the programs at the first position in the historical viewing record.
2. The method of claim 1, wherein the obtaining a preset time period and a time-sharing label weight corresponding to a preset label according to the historical viewing record comprises:
extracting the programs corresponding to the preset time period from the historical viewing records to obtain a time-sharing viewing sequence;
extracting the program corresponding to the preset label from the time-sharing viewing sequence to obtain a time-sharing viewing sequence of the time-sharing viewing label;
and dividing the total number of the programs in the time-sharing and sub-label viewing sequence by the total number of the programs in the time-sharing and sub-label viewing sequence to obtain the weight of the time-sharing and sub-label.
3. The content recommendation method according to claim 1, wherein the obtaining of the program recommendation list corresponding to the preset tag and assigning values to programs in the program recommendation list to obtain recommendation assignment values comprises:
acquiring a public program recommendation list;
extracting all programs corresponding to the preset labels from the public program recommendation list to obtain the program recommendation list corresponding to each preset label;
and assigning the programs in the program recommendation list according to a second preset rule to obtain the recommendation assignment corresponding to each program in the public program recommendation list.
4. A content recommendation system, comprising:
the historical watching record acquisition module is used for acquiring the historical watching record of the user;
the time-sharing label weight obtaining module is used for obtaining a preset time period and time-sharing label weights corresponding to preset labels according to the historical watching records, wherein the preset labels represent program categories;
the influence factor acquisition module is used for acquiring an adding factor and an attenuation factor corresponding to the preset label according to the historical watching record, wherein the adding factor comprises a full-time adding factor and a time-sharing adding factor, and the attenuation factor comprises a full-time attenuation factor and a time-sharing attenuation factor;
the recommended program assignment module is used for acquiring a program recommendation list corresponding to the preset label and assigning the programs in the program recommendation list to obtain a recommendation assignment;
the recommended content obtaining module is used for obtaining recommended content according to the time-sharing tagging weight, the adding factor, the attenuation factor and the recommended assignment;
the influence factor acquisition module comprises:
the program score acquisition unit is used for assigning values to programs in the historical viewing records according to a first preset rule to obtain a program score corresponding to each program, wherein the assignment of the first preset rule to the programs in the historical viewing records comprises incremental assignment of the programs in the historical viewing records according to the time sequence and with preset tolerance;
the system comprises a label-divided viewing sequence acquisition unit, a label-divided viewing sequence acquisition unit and a label-divided viewing sequence acquisition unit, wherein the label-divided viewing sequence acquisition unit is used for extracting programs corresponding to preset labels from historical viewing records to obtain a label-divided viewing sequence;
the full-time influence factor acquisition unit is used for calculating and obtaining a full-time adding and dividing factor and a full-time attenuation factor corresponding to the preset label according to the program value and the view sequence of the sub-label;
the time-sharing viewing sequence acquisition unit is used for extracting programs corresponding to a preset time period from the historical viewing records to obtain a time-sharing viewing sequence;
the time-sharing tag viewing sequence acquisition unit is used for extracting programs corresponding to preset tags from the time-sharing viewing sequence to obtain a time-sharing tag viewing sequence;
the time-sharing influence factor acquisition unit is used for calculating and obtaining a time-sharing adding and dividing factor and a time-sharing attenuation factor corresponding to a preset label according to the program score, the time-sharing viewing sequence and the time-sharing label viewing sequence;
the full-time influence factor acquisition unit includes:
the sub-label program score sum calculating sub-unit is used for calculating the sum of the program scores corresponding to the programs in the sub-label viewing sequence to obtain the sum of the score of the sub-label programs;
the full-time attenuation factor calculating subunit is configured to calculate, based on a first preset calculation rule, a full-time attenuation factor corresponding to a preset tag according to the score sum of the sub-tag programs, the total number of the programs in the sub-tag viewing sequence, the total number of the programs in the historical viewing record, and a first preset attenuation coefficient, where the first preset calculation rule is as follows:
ai＝1+d1*(nL*Num(ni)-Sum(ni))；
wherein, aiFor a full time attenuation factor, d, corresponding to a preset label i1Is a first predetermined attenuation system, nLFor the total number of programs in the historical viewing record, Num (n)i) For the total number of programs in the tab view, Sum (n)i) Is the sum of the scores of the sub-label programs;
the calculation formula of the full-time addend factor is as follows:
bi＝a1*N
wherein, a1Is a first predetermined addendum coefficient, biAdding a score factor for the preset label i;
the N value obtaining subunit is used for obtaining the N value when preset labels corresponding to the continuous N programs in the historical viewing record are the same, wherein N is more than or equal to 2;
the full-time bonus factor acquisition subunit is used for multiplying the N value by a first preset bonus coefficient to obtain a full-time bonus factor corresponding to the preset label;
the time-sharing influence factor acquisition unit includes:
the time-sharing tag program score sum obtaining subunit is used for calculating the sum of the program scores corresponding to the programs in the time-sharing tag watching sequence to obtain the time-sharing tag program score sum;
the time-sharing attenuation factor calculating subunit is configured to calculate, based on a second preset calculation rule, to obtain a time-sharing attenuation factor corresponding to the preset tag according to the time-sharing tag program score sum, the total number of programs in the view receiving sequence of the time-sharing tag, the total number of programs in the historical viewing record, and a second preset attenuation coefficient, where the second preset calculation rule is as follows:
at，i＝1+d2*(nL*Num(nt，i)-Sum(nt，i))；
wherein, at，iTime-sharing attenuation factor, d2For the second predetermined attenuation coefficient, Num (n)t，i) For the total number of programs in the time-division label-receiving view sequence, Sum (n)t，i) Is the sum of the time-sharing label program scores;
the calculation formula of the time-sharing bonus factor corresponding to the preset label is as follows:
bt，i＝a2*M
wherein, bt，iIs a time-sharing bonus factor, a, corresponding to a preset label i2A second preset bonus coefficient;
the M value obtaining subunit is used for obtaining the M value when the preset labels corresponding to the continuous M programs in the time-sharing viewing sequence are the same, wherein M is more than or equal to 2;
the time-sharing bonus factor acquisition subunit is used for multiplying the M value by a second preset bonus coefficient to obtain a time-sharing bonus factor corresponding to the preset label;
the recommended content acquisition module includes:
the recommendation score calculating unit is configured to perform integral operation on the time division label weight, the scoring factor, the attenuation factor and the recommendation assignment based on a third preset calculation rule to obtain a recommendation score of a program in the public program recommendation list, where the third preset calculation rule is:
wherein n ispIndicating the number of occurrences of program p, n, in all public recommendation listsx，pNumber of preset tags representing program p, i represents preset tag i, t represents preset time period t, at，iRepresents a time-sharing attenuation factor corresponding to a preset label i, aiRepresents the full-time attenuation factor corresponding to the preset label i, bt，iRepresenting the time-sharing and scoring factor corresponding to the preset label i, biRepresenting the full time bonus factor corresponding to the preset label i, c representing the preset bonus item, wt，iRepresenting time-of-use sub-label weights, Dx，i，pRepresenting recommendation assignment of the program p obtained by screening by adopting a preset algorithm x in a program recommendation list corresponding to a preset label i;
a recommended content obtaining unit, configured to correspondingly arrange the programs in the public program recommendation list according to a recommendation score from high to low to obtain recommended content, where the preset bonus item c includes a first preset bonus value and a second preset bonus value, and when one preset tag corresponding to the program p is the same as one preset tag corresponding to a program at a first position in the history viewing record, c is the first preset bonus value; and c is a second preset added value when the preset label corresponding to the program p is different from all the preset labels corresponding to the programs at the first position in the historical viewing record.
5. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the content recommendation method according to any one of claims 1 to 3 when executing the computer program.
6. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the content recommendation method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810032233.8A CN108363730B (en) | 2018-01-12 | 2018-01-12 | Content recommendation method, system and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810032233.8A CN108363730B (en) | 2018-01-12 | 2018-01-12 | Content recommendation method, system and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108363730A CN108363730A (en) | 2018-08-03 |
| CN108363730B true CN108363730B (en) | 2020-11-06 |
Family
ID=63006105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810032233.8A Active CN108363730B (en) | 2018-01-12 | 2018-01-12 | Content recommendation method, system and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108363730B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110309358A (en) * | 2018-03-27 | 2019-10-08 | 京东方科技集团股份有限公司 | A kind of resource recommendation method and system |
| CN110598047A (en) * | 2019-08-22 | 2019-12-20 | 优地网络有限公司 | Movie and television information recommendation method and device, electronic equipment and storage medium |
| CN112052378B (en) * | 2019-10-15 | 2021-09-21 | 河南紫联物联网技术有限公司 | Intelligent terminal, and recommendation method and system for intelligent home application |
| CN114302182B (en) * | 2021-12-28 | 2023-10-20 | 未来电视有限公司 | Push method, device, equipment and storage medium for television programs |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957969A (en) * | 2012-05-18 | 2013-03-06 | 华东师范大学 | Device and method for recommending program to IPTV (Internet protocol television) terminal user |
| CN104102648A (en) * | 2013-04-07 | 2014-10-15 | 腾讯科技（深圳）有限公司 | User behavior data based interest recommending method and device |
| CN104216885A (en) * | 2013-05-29 | 2014-12-17 | 酷盛(天津）科技有限公司 | Recommending system and method with static and dynamic recommending reasons automatically combined |
| CN105898421A (en) * | 2015-01-16 | 2016-08-24 | 何湘 | Network film and television program intelligent recommendation method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090210907A1 (en) * | 2008-02-14 | 2009-08-20 | At&T Knowledge Ventures, L.P. | Method and system for recommending multimedia content |
| US7991757B2 (en) * | 2008-08-12 | 2011-08-02 | Amazon Technologies, Inc. | System for obtaining recommendations from multiple recommenders |
-
2018
- 2018-01-12 CN CN201810032233.8A patent/CN108363730B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957969A (en) * | 2012-05-18 | 2013-03-06 | 华东师范大学 | Device and method for recommending program to IPTV (Internet protocol television) terminal user |
| CN104102648A (en) * | 2013-04-07 | 2014-10-15 | 腾讯科技（深圳）有限公司 | User behavior data based interest recommending method and device |
| CN104216885A (en) * | 2013-05-29 | 2014-12-17 | 酷盛(天津）科技有限公司 | Recommending system and method with static and dynamic recommending reasons automatically combined |
| CN105898421A (en) * | 2015-01-16 | 2016-08-24 | 何湘 | Network film and television program intelligent recommendation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108363730A (en) | 2018-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108363730B (en) | Content recommendation method, system and terminal equipment | |
| WO2021042826A1 (en) | Video playback completeness prediction method and apparatus | |
| CN105512331B (en) | A kind of video recommendation method and device | |
| CN109657138A (en) | A kind of video recommendation method, device, electronic equipment and storage medium | |
| CN103559206B (en) | A kind of information recommendation method and system | |
| CA3150500C (en) | Uploader matching method and device | |
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| CN100442284C (en) | Search system for providing information of keyword input frequency by category and method thereof | |
| CN106131601A (en) | Video recommendation method and device | |
| CN106294830A (en) | The recommendation method and device of multimedia resource | |
| CN108304512A (en) | A kind of thick sort method of video search engine, device and electronic equipment | |
| CN107832437A (en) | Audio/video method for pushing, device, equipment and storage medium | |
| CN104462594A (en) | Method and device for providing user personalized resource message pushing | |
| CN108363788B (en) | Post intelligent ranking method and device and computer readable storage medium | |
| CN110472154A (en) | A kind of resource supplying method, apparatus, electronic equipment and readable storage medium storing program for executing | |
| CN103631889B (en) | Image recognizing method and device | |
| CN107948752A (en) | Subscribe to main broadcaster's sort method, device and terminal | |
| CN109447273A (en) | Model training method, advertisement recommended method, relevant apparatus, equipment and medium | |
| CN111222931A (en) | Product recommendation method and system | |
| CN106777201B (en) | Method and device for sorting recommended data on search result page | |
| CN111861542B (en) | Method, device, terminal and non-transitory computer readable storage medium for advertising | |
| CN105956061A (en) | Method and device for determining similarity between users | |
| CN110851708B (en) | Negative sample extraction method, device, computer equipment and storage medium | |
| CN111314790A (en) | Video playing record sequencing method and device and electronic equipment | |
| CN109597934B (en) | Method and device for determining click recommendation words, storage medium and electronic equipment |
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 |