CN114971741A - Method and terminal for positioning network literature user loss sections - Google Patents
Method and terminal for positioning network literature user loss sections Download PDFInfo
- Publication number
- CN114971741A CN114971741A CN202210744248.3A CN202210744248A CN114971741A CN 114971741 A CN114971741 A CN 114971741A CN 202210744248 A CN202210744248 A CN 202210744248A CN 114971741 A CN114971741 A CN 114971741A
- Authority
- CN
- China
- Prior art keywords
- user
- chapters
- unit
- chapter
- book
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012552 review Methods 0.000 claims abstract description 64
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 230000002159 abnormal effect Effects 0.000 claims description 24
- 230000006399 behavior Effects 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 abstract description 7
- 230000014759 maintenance of location Effects 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a method and a terminal for positioning network literature user loss chapters, wherein a preset number of chapters are taken as a unit, the unit user loss rate of each unit of a designated book is calculated, and the average unit user loss rate is calculated; judging whether the unit user churn rate of a first unit is larger than the average unit user churn rate of a preset multiplying power or not, and if so, positioning each chapter in the first unit; acquiring and analyzing book review information of each chapter in the first unit to obtain all chapters with the bad review number exceeding a preset threshold value as problem content chapters; through calculation of the user loss rate and analysis of user evaluation, problem content chapters causing a large amount of loss of the user can be effectively and accurately positioned, so that the content of the chapters can be adjusted subsequently or corresponding processing can be carried out to improve the content quality of the novel, and the purpose of improving user retention is achieved.
Description
Technical Field
The invention relates to the technical field of electronic reading, in particular to a method and a terminal for positioning lost chapters of a network literature user.
Background
Chinese has 5 hundred million network users, and the network novels are recorded in millions, and have different advantages and disadvantages. The influence of the content quality of the novel in the operation process of the network literature novel is particularly important. Through a large amount of data analysis, we can conclude that, among the abnormal user loss reasons, the user loss caused by the novel content problem accounts for more than 80% of the analysis data, and other various reasons account for 20% in total. Therefore, in order to improve the retention rate of the user and reduce the loss of the user, the user needs to accurately find out the problem content chapters, correct and adjust the chapter content or perform corresponding processing to improve the content quality of the novel, so as to achieve the purpose of improving the retention rate of the user.
However, how to locate the problem content section is an urgent problem to be solved at present.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method and the terminal for positioning the network literature user loss chapters can accurately position the problem content chapters causing user loss.
In order to solve the technical problems, the invention adopts the technical scheme that:
a method for positioning network literature user loss sections comprises the following steps:
s1, calculating the unit user attrition rate of each unit of the appointed book by taking a preset number of chapters as a unit, and calculating the average unit user attrition rate;
s2, judging whether the unit user churn rate of a first unit is larger than the average unit user churn rate of a preset multiplying power, and if so, positioning each chapter in the first unit;
and S3, acquiring and analyzing the book comment information of each chapter in the first unit, and obtaining all chapters with the bad comment quantity exceeding a preset threshold value as question content chapters.
In order to solve the technical problem, the invention adopts another technical scheme as follows:
a terminal for locating network literature user churn chapters, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, when executing the computer program, implementing the steps of:
s1, calculating the unit user attrition rate of each unit of the appointed book by taking a preset number of chapters as a unit, and calculating the average unit user attrition rate;
s2, judging whether the unit user churn rate of a first unit is larger than the average unit user churn rate of a preset multiplying power, and if so, positioning each chapter in the first unit;
and S3, acquiring and analyzing the book comment information of each chapter in the first unit, and obtaining all chapters with the bad comment quantity exceeding a preset threshold value as question content chapters.
The invention has the beneficial effects that: according to the method, the electronic book calculates the user loss rate by taking the preset number of chapters as a unit, so that a range of the problem chapters causing the user loss is determined, and the chapters with poor user evaluation, namely the problem content chapters causing the user loss, are further screened out according to the user evaluation condition of the chapters in the range, so that the problem content chapters causing the user loss can be effectively and accurately positioned, the content of the chapters can be adjusted or correspondingly processed subsequently, the content quality of the novel can be improved, and the purpose of improving the user retention can be achieved.
Drawings
Fig. 1 is a flowchart of a method for locating network literature user churn sections according to an embodiment of the present invention;
fig. 2 is a structural diagram of a terminal for positioning network literature user churn chapters according to an embodiment of the present invention;
fig. 3 is a detailed flowchart of a method for locating network literature user churn sections according to an embodiment of the present invention;
description of reference numerals:
1. a terminal for positioning network literature user loss chapters; 2. a processor; 3. a memory.
Detailed Description
In order to explain technical contents, achieved objects, and effects of the present invention in detail, the following description is made with reference to the accompanying drawings in combination with the embodiments.
Referring to fig. 1, a method for locating network literature user churn sections includes the steps:
s1, calculating the unit user attrition rate of each unit of the appointed book by taking a preset number of chapters as a unit, and calculating the average unit user attrition rate;
s2, judging whether the unit user churn rate of a first unit is larger than the average unit user churn rate of a preset multiplying power, and if so, positioning each chapter in the first unit;
and S3, acquiring and analyzing the book comment information of each chapter in the first unit, and obtaining all chapters with the bad comment quantity exceeding a preset threshold value as question content chapters.
From the above description, the beneficial effects of the present invention are: according to the method, the electronic book calculates the user loss rate by taking the preset number of chapters as a unit, so that a range of the problem chapters causing the user loss is determined, and the chapters with poor user evaluation, namely the problem content chapters causing the user loss, are further screened out according to the user evaluation condition of the chapters in the range, so that the problem content chapters causing the user loss can be effectively and accurately positioned, the content of the chapters can be adjusted or correspondingly processed subsequently, the content quality of the novel can be improved, and the purpose of improving the user retention can be achieved.
Further, the step S1 is specifically:
s11, calculating the unit user churn rate of each unit by taking a preset number of chapters as a unit aiming at the payment chapters of the appointed book, wherein the calculation of the unit user churn rate specifically comprises the following steps:
rate＝1-[num(n+a)/num(n)]；
wherein a represents a preset number, num (n + a) represents the number of unlocking users of the (n + a) th chapter, num represents the number of unlocking users of the nth chapter, and the first unit comprises the (n + 1) th chapter to the (n + a) th chapter;
and S12, calculating the average value of the unit user loss rate of each unit to obtain the average unit user loss rate.
From the above description, the user loss rate during the period from chapter n +1 to chapter n + a is calculated according to the number of unlocking users in chapter n and the number of unlocking users in chapter n + a, which is direct, reasonable and more accurate.
Further, the step between the step S2 and the step S3 further includes the steps of:
s21, acquiring chapter unlocking information of each lost user in a first unit, and judging whether the lost user is a normal lost user or an abnormal lost user;
judging whether the lost user is a normal lost user or an abnormal lost user specifically comprises the following steps:
judging whether electronic money or reading gift certificates are reserved in the account after the user unlocks the payment section finally, if so, the user is abnormally lost, otherwise, the user is normally lost;
the book rating information in step S3 is the book rating information of all the users who abnormally lose.
From the above description, if there is no user loss caused by remaining electronic money, we consider that the user loss is normal loss, and the user is not very interested in the content of the book, but only in order to consume the electronic money on hand, the part of people is not in the processing range of the scheme, and relatively, the part of lost users who are not used by the remaining electronic money on the account but no longer unlock the subsequent chapters is the object of the analysis research by the scheme, i.e. the abnormal lost user, and we screen the abnormal lost user for the subsequent review analysis of the book, and then determine the chapters of the problem content, thereby reducing the required computing resources and improving the accuracy of positioning the problem content.
Further, the step S3 of obtaining all chapters in which the number of bad comments exceeds the preset threshold as question content chapters specifically includes:
judging whether chapters with the bad comment quantity exceeding a preset threshold exist in the first unit, if so, marking all chapters with the bad comment quantity exceeding the preset threshold as problem content chapters, and otherwise, entering the step S4;
the step S3 is followed by the step of:
s4, acquiring and analyzing reading behavior information of an abnormal loss user, and calculating a first average page turning speed of the abnormal loss user for reading the specified book and a first average operation interval time for reading a next chapter, and a second average page turning speed of each chapter in a first unit for reading and a second average operation interval time for reading the next chapter;
and if the difference value between the second average page turning speed and the first average page turning speed of the chapters exceeds a preset speed threshold value, or the difference value between the second average operation interval time and the first average operation interval time exceeds a preset interval threshold value, marking the chapters as problem content chapters.
As can be seen from the above description, after the analysis is performed according to the good comment and the bad comment of the book comment, the problem content section still cannot be determined, and we can perform the analysis by the user behavior abnormality of the user, thereby identifying the problem content section.
Further, the step S3 is specifically:
s31, acquiring book review information in the first unit, screening the book review information according to a preset keyword library, eliminating invalid book review information in the book review information, and screening out bad reviews in the valid book review information;
and S32, screening all chapters with the number of bad comments exceeding a preset threshold value in the first unit as question content chapters.
As can be seen from the above description, the present invention screens the reviews of the books through the preset keyword library to screen out invalid reviews of the books, such as randomly typed character strings, watered characters, characters without actual meaning or content, etc., leaves valid reviews of the books and screens out bad reviews of the books, and judges whether the chapters are question content chapters according to the number of the bad reviews.
Further, the step between the step S31 and the step S32 further includes the steps of:
s311, judging whether the book review information is valid or bad, and if the book review information is valid or bad, sending the book review information which cannot be identified to manual work for identification.
As can be seen from the above description, for the book comments which cannot be intelligently identified, the book comments can be sent to the manual work for identification.
Further, the step S3 is replaced with:
reading behavior information of an abnormal loss user is obtained and analyzed, and a first average page turning speed of the abnormal loss user for reading the appointed book and a first average operation interval time for reading a next chapter, and a second average page turning speed of each chapter in a first unit for reading and a second average operation interval time for reading the next chapter are calculated;
and if the difference value between the second average page turning speed and the first average page turning speed of the chapters exceeds a preset speed threshold value, or the difference value between the second average operation interval time and the first average operation interval time exceeds a preset interval threshold value, marking the chapters as problem content chapters.
As can be seen from the above description, after the chapters in the first unit are located, the reading behavior of the user can also be directly analyzed, so as to identify the question content chapters.
Further, the step S3 is replaced with:
and acquiring book review information of each chapter in the first unit, performing semantic analysis on the book review information according to a preset keyword library, and marking the chapter corresponding to the first book review as a question content chapter if the semantics of the first book review fall into a preset semantic set.
As can be seen from the above description, after each chapter in the first unit is located, the question content chapter can also be determined by analyzing the semantics of the book review, such as the book review with specific semantics, e.g., the book review with semantics "x is very annoying and book discard".
Further, the preset number is 10.
As can be seen from the above description, the unit of chapter 10 is neater and easier to count.
Referring to fig. 2, a terminal for locating network literature user churn chapters includes a processor, a memory, and a computer program stored in the memory and executable on the processor, where the processor implements the steps when executing the computer program.
The method and the terminal for positioning the network literature user loss sections are suitable for a network reading industry, and are used for finding a scene that a certain book has a large number of readers lost and needs to be traced.
Referring to fig. 1 and fig. 3, a first embodiment of the present invention is:
a method for positioning network literature user loss sections comprises the following steps:
s1, calculating the unit user attrition rate of each unit of the appointed book by taking a preset number of chapters as a unit, and calculating the average unit user attrition rate;
the step S1 specifically includes:
s11, calculating a unit user churn rate of each unit by taking a preset number of chapters as a unit for paying chapters of a specific book, where the calculation of the unit user churn rate specifically includes:
rate＝1-[num(n+a)/num(n)]；
wherein a represents a preset number, num (n + a) represents the number of unlocking users of the (n + a) th chapter, num represents the number of unlocking users of the nth chapter, and the first unit comprises the (n + 1) th chapter to the (n + a) th chapter;
and S12, calculating the average value of the unit user loss rate of each unit to obtain the average unit user loss rate.
In this embodiment, we monitor chapter run-off rate for the explosive book a in the platform. For each chapter 10 of the payment chapter of book a, a unit user churn rate [1- (chapter N +10 unlock people/chapter N unlock people) ] and an average unit user churn rate are calculated.
In this embodiment, the preset number is 10, and in other equivalent embodiments, the preset number may also be other values, which are set according to actual requirements.
S2, judging whether the unit user churn rate of a first unit is larger than the average unit user churn rate of a preset multiplying power, and if so, positioning each chapter in the first unit;
in this embodiment, the unit user loss rate of each pay chapter of 10 chapters is compared with the average unit user loss rate, and if the unit user loss rate of a certain chapter is more than 2 times of the average unit user loss rate, an early warning is given to obtain a large loss chapter range.
In this embodiment, in the process of monitoring the average payment rate and the average loss rate of the book a, the program finds that the user has a large loss between chapters 105 and 114 (payment chapters) of the book, so the data abnormal chapter determined by us is chapter 105 and 115 of the book.
The method between the step S2 and the step S3 further comprises the steps of:
s21, acquiring chapter unlocking information of each lost user in a first unit, and judging whether the lost user is a normal lost user or an abnormal lost user;
judging whether the lost user is a normal lost user or an abnormal lost user specifically comprises the following steps:
judging whether electronic money or reading gift certificates are reserved in the account after the user unlocks the payment section finally, if so, the user is abnormally lost, otherwise, the user is normally lost;
the book rating information in step S3 is the book rating information of all the users who abnormally lose.
In this embodiment, the criterion of the normal loss of the user is that after the user unlocks the last reading chapter, whether the user has remaining money or gift certificates in the account, and if the user has no loss due to remaining electronic money, the user is deemed to be normal loss, which is not very interested in the content of the book, but only in order to consume the electronic money on the hand, and this part of people is not in the processing category of the scheme. In contrast, the part of the lost users who still have the remaining electronic money on the account and do not unlock the subsequent chapters is the object of the scheme to be analyzed and researched, namely, the abnormal lost users.
When the range of the problem content chapter is located (large loss chapter range), due to the fact that the number of the book chapters is large and the number of readers is large, under the condition that the characteristic that a large number of users consume electronic money or gift certificates in a certain unit does not exist, the influence of normal loss users on the range location is relatively small, but when the problem chapter needs to be accurately located in the large loss chapter range, the interference of normal loss should be eliminated. Meanwhile, due to the fact that book review is needed subsequently or user behaviors are possibly analyzed, the calculation resources occupied by the subsequent steps can be effectively reduced by removing users who are lost normally.
In this embodiment, for chapter 10 of chapters 105-115, the system performs data analysis on the user who has read chapter 10, and screens out the user who runs normally by determining whether there is digital currency remaining but no longer paying for unlocking the chapters, and the screened out user who runs abnormally enters the book review analysis link of the next link.
S3, acquiring and analyzing book review information of each chapter in the first unit, and obtaining all chapters with the bad review number exceeding a preset threshold value as question content chapters;
the step S3 specifically includes:
s31, acquiring book review information in the first unit, screening the book review information according to a preset keyword library, eliminating invalid book review information in the book review information, and screening out bad reviews in the valid book review information;
s311, judging whether the book review information is not identified to be effective or bad, and if the book review information is not identified to be effective or bad, sending the book review information which is not identified to a worker for identification;
and S32, screening all chapters with the number of bad comments exceeding a preset threshold value in the first unit as question content chapters.
In the embodiment, the keyword library which is manually collected and continuously supplemented is used for book review screening, invalid book reviews are eliminated, and the intelligent classification of the valid book reviews into good reviews and poor reviews is carried out. And for part of the book reviews which cannot be intelligently identified and divided, manual analysis can be carried out. The method for distinguishing good comments from bad comments is a method for continuously enriching comment libraries through artificial learning by utilizing a keyword library and analyzing and reading the comment contents to classify comments (the comments can be classified by reading according to Taobao goods comments). If some chapters have a large number of bad scores (exceeding a manually set threshold), then those chapters are located as problem content chapters.
Referring to fig. 3, the second embodiment of the present invention is:
a method for locating sections lost by network literature users, which is different from the first embodiment in that, in the step S3, the step of obtaining all sections in which the number of bad comments exceeds a preset threshold is specifically a question content section:
and judging whether chapters with the bad comment quantity exceeding a preset threshold exist in the first unit, if so, marking all chapters with the bad comment quantity exceeding the preset threshold as problem content chapters, and otherwise, entering the step S4.
S4, acquiring and analyzing reading behavior information of an abnormal loss user, and calculating a first average page turning speed of the abnormal loss user for reading the specified book and a first average operation interval time for reading a next chapter, and a second average page turning speed of each chapter in a first unit for reading and a second average operation interval time for reading the next chapter;
and if the difference value between the second average page turning speed and the first average page turning speed of the chapters exceeds a preset speed threshold value, or the difference value between the second average operation interval time and the first average operation interval time exceeds a preset interval threshold value, marking the chapters as problem content chapters.
In this embodiment, if there are no valid chapters or a large number of badly rated chapters within the range of the checked large-loss chapters, the user behavior is determined. By reading an APP or a page program and the like in advance, reading behavior data of a user is counted when the user reads, such as an average page turning speed and an average operation interval time of the user reading a next chapter. We compare the chapter paging speed and the operation interval time within the investigation range. And if the page turning speed of a user in a certain chapter is greater than or less than the average page turning speed of the whole book and exceeds a preset speed threshold, or the time interval between the user entering the next chapter is greater than or less than the average time interval between the user entering the next chapter and the whole book and exceeds a preset interval threshold, locating the chapter as a problem content chapter.
The preset speed threshold and the preset interval threshold may be set as percentages of the average page turning speed or the average operation interval time corresponding to the whole book, or may be specific numerical values. In actual use, a normal numerical range can be specified according to the average page turning speed or the average operation interval time of the whole book, and if the range is exceeded, the chapter is positioned as a question content chapter.
The problem content chapters determined through the flow of the scheme are specific chapters, and after the problem chapter data is received subsequently, the chapter range with more concentrated data is clear actually. If further needs exist, the scheme of the invention also achieves the effect of further reducing the scope of problem sections.
The third embodiment of the invention is as follows:
a method for locating network literature user churn chapters, which is different from the first embodiment in that step S3 is replaced by:
reading behavior information of an abnormal loss user is obtained and analyzed, and a first average page turning speed of the abnormal loss user for reading the appointed book and a first average operation interval time for reading a next chapter, and a second average page turning speed of each chapter in a first unit for reading and a second average operation interval time for reading the next chapter are calculated;
and if the difference value between the second average page turning speed and the first average page turning speed of the chapters exceeds a preset speed threshold value, or the difference value between the second average operation interval time and the first average operation interval time exceeds a preset interval threshold value, marking the chapters as problem content chapters.
The fourth embodiment of the invention is as follows:
a method for locating network literature user churn chapters, which is different from the first embodiment in that step S3 is replaced by:
and acquiring book review information of each chapter in the first unit, performing semantic analysis on the book review information according to a preset keyword library, and marking the chapter corresponding to the first book review as a question content chapter if the semantics of the first book review fall into a preset semantic set.
In this embodiment, after the book review information of each chapter in the first unit is acquired, semantic analysis is performed on each book review through the preset keyword library to determine whether any semantic in the preset semantic set is satisfied, and if the satisfied book review exists, the chapter corresponding to the book review is marked as a problem content chapter. The semantics of the preset semantic set include ". x.is annoying and leaves a book", and the like, and the content of the chapter has a negative emotion or indicates the intention to leave a book.
Referring to fig. 2, a fifth embodiment of the present invention is:
a terminal 1 for locating network literature user churn chapters comprises a processor 2, a memory 3 and a computer program stored in the memory 3 and capable of running on the processor 2, wherein the processor 2 realizes the steps in the first embodiment when executing the computer program.
In summary, the method and the terminal for positioning the network literature user loss chapters provided by the present invention determine a range in which the problem chapters causing the user loss exist by calculating the user loss rate through the electronic book with a preset number of chapters as a unit, further screen out chapters with poor user evaluation according to the user evaluation condition of the chapters in the range, and screen out chapters with abnormal behavior according to the user behavior data, that is, the problem content chapters causing the user loss, and can effectively and accurately position the problem content chapters causing the user loss in order to subsequently adjust the chapter content or perform corresponding processing to improve the content quality of the novel, thereby achieving the purpose of improving the user retention. And the identification of normal loss users and abnormal loss users can be effectively realized according to the user account condition, so that the positioning of problem content chapters is more efficient and accurate.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method for positioning network literature user loss chapters is characterized by comprising the following steps:
s1, calculating the unit user attrition rate of each unit of the appointed book by taking a preset number of chapters as a unit, and calculating the average unit user attrition rate;
s2, judging whether the unit user churn rate of a first unit is larger than the average unit user churn rate of a preset multiplying power, and if so, positioning each chapter in the first unit;
and S3, acquiring and analyzing the book comment information of each chapter in the first unit, and obtaining all chapters with the bad comment quantity exceeding a preset threshold value as question content chapters.
2. The method for locating sections lost by network literature users according to claim 1, wherein the step S1 specifically comprises:
s11, calculating the unit user churn rate of each unit by taking a preset number of chapters as a unit aiming at the payment chapters of the appointed book, wherein the calculation of the unit user churn rate specifically comprises the following steps:
rate＝1-[num(n+a)/num(n)]；
wherein a represents a preset number, num (n + a) represents the number of unlocking users of the (n + a) th chapter, num represents the number of unlocking users of the nth chapter, and the first unit comprises the (n + 1) th chapter to the (n + a) th chapter;
and S12, calculating the average value of the unit user loss rate of each unit to obtain the average unit user loss rate.
3. The method for locating network literature user churn chapters according to claim 1, wherein the step between the step S2 and the step S3 further comprises the steps of:
s21, acquiring chapter unlocking information of each lost user in a first unit, and judging whether the lost user is a normal lost user or an abnormal lost user;
judging whether the lost user is a normal lost user or an abnormal lost user specifically comprises the following steps:
judging whether electronic money or reading gift certificates are reserved in the account after the user unlocks the payment section finally, if so, the user is abnormally lost, otherwise, the user is normally lost;
the book rating information in step S3 is the book rating information of all the users who abnormally lose.
4. The method according to claim 3, wherein the step S3 of obtaining all chapters with the number of bad comments exceeding a preset threshold is specifically:
judging whether chapters with the bad comment quantity exceeding a preset threshold exist in the first unit, if so, marking all chapters with the bad comment quantity exceeding the preset threshold as problem content chapters, and otherwise, entering the step S4;
the step S3 is followed by the step of:
s4, acquiring and analyzing reading behavior information of an abnormal loss user, and calculating a first average page turning speed of the abnormal loss user for reading the specified book and a first average operation interval time for reading a next chapter, and a second average page turning speed of each chapter in a first unit for reading and a second average operation interval time for reading the next chapter;
and if the difference value between the second average page turning speed and the first average page turning speed of the chapters exceeds a preset speed threshold value, or the difference value between the second average operation interval time and the first average operation interval time exceeds a preset interval threshold value, marking the chapters as problem content chapters.
5. The method of claim 1, wherein the step S3 includes:
s31, acquiring book review information in the first unit, screening the book review information according to a preset keyword library, eliminating invalid book review information in the book review information, and screening out bad reviews in the valid book review information;
and S32, screening all chapters with the number of bad comments exceeding a preset threshold value in the first unit as problem content chapters.
6. The method of claim 5, wherein between the step S31 and the step S32, the method further comprises the steps of:
s311, judging whether the book review information is valid or bad, and if the book review information is valid or bad, sending the book review information which cannot be identified to manual work for identification.
7. The method for locating sections lost by network literature users according to claim 1, wherein the step S3 is replaced by:
reading behavior information of an abnormal loss user is obtained and analyzed, and a first average page turning speed of the abnormal loss user for reading the appointed book and a first average operation interval time for reading a next chapter, and a second average page turning speed of each chapter in a first unit for reading and a second average operation interval time for reading the next chapter are calculated;
and if the difference value between the second average page turning speed and the first average page turning speed of the chapters exceeds a preset speed threshold value, or the difference value between the second average operation interval time and the first average operation interval time exceeds a preset interval threshold value, marking the chapters as problem content chapters.
8. The method for locating network literature user churn chapters according to claim 1, wherein the step S3 is replaced by:
and acquiring book review information of each chapter in the first unit, performing semantic analysis on the book review information according to a preset keyword library, and marking the chapter corresponding to the first book review as a question content chapter if the semantics of the first book review fall into a preset semantic set.
9. The method of claim 1, wherein the predetermined number is 10.
10. A terminal for locating network literature user churn chapters, comprising a processor, a memory and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program to implement the steps of any one of the methods for locating network literature user churn chapters of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210744248.3A CN114971741B (en) | 2022-06-27 | 2022-06-27 | Method and terminal for positioning loss chapters of network literature users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210744248.3A CN114971741B (en) | 2022-06-27 | 2022-06-27 | Method and terminal for positioning loss chapters of network literature users |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114971741A true CN114971741A (en) | 2022-08-30 |
| CN114971741B CN114971741B (en) | 2024-09-06 |
Family
ID=82965050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210744248.3A Active CN114971741B (en) | 2022-06-27 | 2022-06-27 | Method and terminal for positioning loss chapters of network literature users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114971741B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040088730A1 (en) * | 2002-11-01 | 2004-05-06 | Srividya Gopalan | System and method for maximizing license utilization and minimizing churn rate based on zero-reject policy for video distribution |
| US8744898B1 (en) * | 2010-11-12 | 2014-06-03 | Adobe Systems Incorporated | Systems and methods for user churn reporting based on engagement metrics |
| CN110781431A (en) * | 2019-09-29 | 2020-02-11 | 上海连尚网络科技有限公司 | Method and equipment for providing novel information |
| CN114116822A (en) * | 2021-11-29 | 2022-03-01 | 北京得间科技有限公司 | Information push method, terminal and storage medium |
-
2022
- 2022-06-27 CN CN202210744248.3A patent/CN114971741B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040088730A1 (en) * | 2002-11-01 | 2004-05-06 | Srividya Gopalan | System and method for maximizing license utilization and minimizing churn rate based on zero-reject policy for video distribution |
| US8744898B1 (en) * | 2010-11-12 | 2014-06-03 | Adobe Systems Incorporated | Systems and methods for user churn reporting based on engagement metrics |
| CN110781431A (en) * | 2019-09-29 | 2020-02-11 | 上海连尚网络科技有限公司 | Method and equipment for providing novel information |
| CN114116822A (en) * | 2021-11-29 | 2022-03-01 | 北京得间科技有限公司 | Information push method, terminal and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 戴和忠: "基于用户阅读行为的图书自动评测算法", 浙江大学学报(工学版), vol. 47, no. 10, 15 October 2013 (2013-10-15), pages 1753 - 1757 * |
| 李明: "网络书店图书差评分析及其启示", 图书馆杂志, vol. 31, no. 09, 15 September 2012 (2012-09-15), pages 58 - 94 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114971741B (en) | 2024-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10482093B2 (en) | Data mining method | |
| CN109977403B (en) | Malicious comment information identification method and device | |
| CN108550054B (en) | Content quality evaluation method, device, equipment and medium | |
| TW201732662A (en) | Method and apparatus for establishing data identification model | |
| CN111523996A (en) | Approval method and system | |
| CN110263155B (en) | Data classification method, and training method and system of data classification model | |
| CN115205866A (en) | Block chain-based scientific and technological paper big data plagiarism detection method and system | |
| CN108462624B (en) | Junk mail identification method and device and electronic equipment | |
| CN111061998A (en) | Analysis model and method for economic measurement | |
| CN115631494A (en) | Financial data processing method, system, electronic device and storage medium | |
| CN110866831A (en) | Asset activity level determination method and device and server | |
| CN109409091B (en) | Method, device and equipment for detecting Web page and computer storage medium | |
| CN114971741A (en) | Method and terminal for positioning network literature user loss sections | |
| CN106682516A (en) | Detection method, detection device and server of application programs | |
| CN117349502A (en) | Operation and maintenance data query analysis method and system based on internet data center | |
| CN111274949A (en) | Structural analysis-based blood disease white blood cell scatter diagram similarity analysis method | |
| CN111324463A (en) | Engineering file label clearing method, system, device and storage medium | |
| CN118133812B (en) | Information analysis method and system based on large language model | |
| CN114567482A (en) | Alarm classification method and device, electronic equipment and storage medium | |
| CN113342622A (en) | Operation behavior auditing method and device and storage medium | |
| CN113962216A (en) | Text processing method and device, electronic equipment and readable storage medium | |
| CN118586386A (en) | Identification method, device and readable medium for certificate-regulating data title line based on part-of-speech comparison | |
| CN117789713B (en) | Health literacy investigation quality control method and system based on voice recognition | |
| CN113095892B (en) | Data processing method and device | |
| CN109657045B (en) | Method and device for acquiring vocabulary emotion value, storage medium and processor |
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 |