WO2019184582A1 - 电子书消费需求的确定方法、电子设备及计算机存储介质 - Google Patents

电子书消费需求的确定方法、电子设备及计算机存储介质 Download PDF

Info

Publication number
WO2019184582A1
WO2019184582A1 PCT/CN2019/073492 CN2019073492W WO2019184582A1 WO 2019184582 A1 WO2019184582 A1 WO 2019184582A1 CN 2019073492 W CN2019073492 W CN 2019073492W WO 2019184582 A1 WO2019184582 A1 WO 2019184582A1
Authority
WO
WIPO (PCT)
Prior art keywords
interaction
book
duration
time
preset
Prior art date
Application number
PCT/CN2019/073492
Other languages
English (en)
French (fr)
Inventor
王海璐
柳燕煌
杨贝贝
王雪梅
Original Assignee
掌阅科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 掌阅科技股份有限公司 filed Critical 掌阅科技股份有限公司
Publication of WO2019184582A1 publication Critical patent/WO2019184582A1/zh

Links

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
    • 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

Definitions

  • the present disclosure relates to the field of computers, and in particular, to a method for determining a consumer demand for an electronic book, an electronic device, and a computer storage medium.
  • the inventor found that simply adding the e-books that the user clicks, downloads, or read to the demand list is too simple and extensive, and does not consider the relationship between the e-books and the reading users.
  • the depth of interaction makes it easy to add e-books that users click on by mistakes, or e-books that users are completely uninterested and only close for a few seconds to join the demand list, so that the final consumer demand cannot truly reflect the potential consumption power of users.
  • the conversion rate of the active content pushed according to this is not high, and the effect of improving the user's reading cannot be achieved.
  • the present disclosure has been made in order to provide a method for determining an e-book consumption demand that overcomes the above problems or at least partially solves the above problems, an electronic device, and a computer storage medium.
  • a method for determining an e-book consumption demand including:
  • the consumption demand information of the reading user is determined.
  • an electronic device includes: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface complete communication with each other through a communication bus; the memory is configured to store at least one Executable instructions that cause the processor to:
  • the consumption demand information of the reading user is determined according to the selected depth interactive e-book and the consumption amount of the corresponding deep interactive e-book.
  • a non-transitory computer readable storage medium having stored therein at least one executable instruction that causes a processor to perform the following operations :
  • the consumption demand information of the reading user is determined according to the selected depth interactive e-book and the consumption amount of the corresponding deep interactive e-book.
  • a computer program product comprising a computing program stored on the non-transitory computer readable storage medium described above.
  • the electronic device, and the computer storage medium the time interval between two adjacent interaction operations in each interaction operation performed by the reading user for each electronic book is determined.
  • Read the interaction duration of the user for each e-book and filter the deep interactive e-book corresponding to the reading user according to the interaction duration, and finally calculate the consumer demand according to the deep interactive e-book.
  • the present disclosure quantifies the user based on a series of interaction operations generated by the user for the e-book, taking the time interval between the interaction operations as a consideration dimension.
  • the depth of interaction with the e-book and thus reflect the user's desire to read the e-book, and further combine the e-book consumption amount to convert the user's desire for e-books into the user's potential consumption demand for the book. This increases the value of operational data and increases the conversion rate of operational activities.
  • FIG. 1 is a flowchart of a method for determining an e-book consumption requirement according to Embodiment 1 of the present disclosure
  • FIG. 2 is a flowchart of a method for determining an e-book consumption requirement provided by Embodiment 2 of the present disclosure
  • FIG. 3 is a schematic structural diagram of an electronic device according to Embodiment 4 of the present disclosure.
  • Embodiment 1 is a flow chart showing a method for determining an e-book consumption demand provided by Embodiment 1 of the present disclosure. As shown in Figure 1, the method includes the following steps:
  • Step S110 Acquire an e-book that interacts with the reading user, and determine, for each e-book obtained, an operation time stamp of each interaction operation performed by the reading user for the e-book.
  • each e-book included in the reading history record of the reading user can be used as an e-book that interacts with the reading user.
  • the historical reading record is used to record the reading behavior of the reading user and its corresponding e-book information.
  • reading behaviors include: downloading, clicking, reading, opening, and paying (here refers to the payment behavior of the first part of the serialized book, rather than the full payment behavior) and other possible behaviors. This approach avoids omissions and ensures that the most comprehensive information is collected.
  • the e-book interacting with the reading user may be determined by: determining, for each e-book included in the historical reading record of the reading user, whether the reading user has performed a preset interaction operation on the e-book. If so, it is identified as an e-book that interacts with the reading user.
  • the preset interaction operation may only include a specific type of operation that can reflect the user's initial interest in the e-book, such as a search operation, a download operation, an entry book detail page operation, a view directory operation, and the like. This method can directly obtain an e-book that the user may be interested in to improve the subsequent processing efficiency.
  • an operation time stamp for reading each interactive operation performed by the user for the e-book is determined. Specifically, for each e-book, when the reading user performs a plurality of interactive operations on the e-book, the operation time stamps of each inter-operation are respectively acquired.
  • the operation timestamp refers to the system time corresponding to the execution of the interaction. It can be seen that the operation time stamp of each interaction operation can determine the operation sequence of each interaction operation.
  • Step S120 Determine the interaction duration of the reading user for the e-book according to the time interval between two adjacent interaction operations in each interaction operation.
  • the time interval between two adjacent interaction operations in each interaction operation is further determined.
  • two adjacent interaction operations in each interaction operation include two adjacent interaction operations in each interaction operation, that is, each adjacent two interaction operations are considered.
  • the two adjacent interaction operations in each interaction operation only include several groups of adjacent two interaction operations that meet the preset rule. For example, only two adjacent interaction operations whose time interval is not greater than a preset value are included, thereby filtering out interaction operations with excessive time intervals. As another example, only two adjacent interactions within a preset time period (eg, the last three months) are included, thereby eliminating excessively long interactions.
  • the disclosure does not limit the specific content and implementation manner of the preset rules, as long as the user requirements can be more truly reflected.
  • the time between two adjacent interaction operations in each interaction operation may be The interval is subjected to preset processing, such as cumulative summation, to obtain the interaction duration of the reading user for the e-book.
  • preset processing such as cumulative summation
  • the inventor has found in the process of implementing the present disclosure that in addition to the instantaneously completed interactive operation, the inventor may also include reading, writing a comment, and the like, which have a long duration operation, and the operation can also reflect the user and the electronic
  • the real interaction process of the book therefore, in order to more accurately reflect the degree of interaction between the user and the e-book, optionally, in this step, when it is determined that each inter-operation includes a preset continuous class operation, further The interaction duration of the reading user for the e-book is determined in conjunction with the duration of the operation corresponding to the preset continuous class operation.
  • the continuous class operation refers to: operations that are completed for a period of time, such as reading operations, writing comment operations, etc., for operations that are completed instantaneously, such as reading operations and writing comment operations. .
  • Step S130 Filter the deep interactive e-book corresponding to the reading user from the obtained e-book according to the interaction duration of the read e-book.
  • the interaction duration is obtained according to the interactive operation of the reading user for the e-book, the depth of interaction between the reading user and the e-book can be reflected. Therefore, the deep interactive e-book corresponding to the reading user can be selected according to the interaction duration.
  • all the e-books may be sorted according to the interaction duration of each e-book, and the first plurality of e-books are selected as the deep interactive e-book according to the sorting result.
  • the deep interactive e-book corresponding to the reading user may be filtered according to a preset screening rule.
  • the screening rules may be determined according to factors such as interaction duration and operation type. The above two methods can be used alone or in combination.
  • the deep interactive e-book described in this embodiment refers to an e-book in which the index data meets a predetermined threshold in the process of measuring the quantifiable index data.
  • the predetermined threshold may be a pre-order number TOP N (positive integer) sorted according to the interaction duration, and the e-book sorted in the first N is a deep interactive e-book; or, the predetermined threshold may be It is a preset interaction duration, and the e-book that is greater than or equal to the interaction duration is a deep interactive e-book; or, when the quantifiable indicator data is the number of times generated by the specific operation type, the predetermined threshold may be a specifically set number of times M (a positive integer) For example, when the user enters the purchase interface of an e-book twice (but not purchased), it indicates that the user has a certain interest in reading the e-book but hesitates to pay for the purchase, so the e-book can be identified as
  • Step S140 Determine the consumption demand information of the reading user according to the collected depth interactive e-book and the consumption amount of the corresponding deep interactive e-book.
  • the amount of consumption of the deep interactive e-book mainly refers to the amount of the unpaid portion corresponding to the e-book.
  • the entire purchase of an e-book refers to the entire price of the e-book; for an e-book purchased in accordance with the chapter, the price of the unpurchased chapter in the e-book.
  • the consumption demand information of the reading user may be determined according to the sum of the consumption amounts of the respective deep interactive e-books.
  • the reading user is determined according to the time interval between two adjacent interaction operations in each interaction operation performed by the reading user for each e-book.
  • the interaction time of each e-book is selected, and the deep interactive e-book corresponding to the reading user is selected according to the interaction duration, and finally the consumption demand is calculated according to the deep interactive e-book.
  • the present disclosure quantifies the user based on a series of interaction operations generated by the user for the e-book, taking the time interval between the interaction operations as a consideration dimension.
  • the depth of interaction with the e-book and thus reflect the user's desire to read the e-book, and further combine the e-book consumption amount to convert the user's desire for e-books into the user's potential consumption demand for the book. This increases the value of operational data and increases the conversion rate of operational activities.
  • FIG. 2 is a flow chart showing a method for determining an e-book consumption demand provided by Embodiment 2 of the present disclosure. As shown in FIG. 2, the method includes the following steps:
  • Step S210 Acquire an e-book that interacts with the reading user.
  • the step may be implemented by: determining each e-book included in the historical reading record of the reading user; and performing culling processing on each e-book included in the historical reading record of the reading user according to the preset culling rule; The e-book remaining after performing the culling process is determined to be an e-book that interacts with the reading user.
  • each e-book included in the historical reading record of the reading user may be searched, or only the reading user may perform the preset interactive operation.
  • the preset culling rule each of the found e-books is culled.
  • the default culling rule is used to eliminate e-books that are not related to consumer demand, that is, e-books that do not contribute to the calculation of consumer demand, including: e-books that are not of interest to users (ie, insufficient depth of interaction) and no potential consumption space.
  • At least one of an e-book (such as a free or fully paid e-book).
  • the preset culling rule may include at least one of the following:
  • Step S220 Determine, for each acquired e-book, an operation time stamp of each interaction operation performed by the reading user for the electronic book.
  • the history reading record includes an operation record of reading each interaction operation performed by the user for each electronic book, where each operation record includes an operation time stamp when performing the interaction operation and an operation object of the interaction operation. (ie e-book). Accordingly, by analyzing each operation record, it is possible to determine an operation time stamp for reading each interactive operation performed by the user for the electronic book for each of the acquired electronic books.
  • Step S230 Determine the interaction duration of the reading user for the e-book according to the time interval between two adjacent interaction operations in each interaction operation.
  • the e-book obtained in step S210 may include an e-book that the user is not interested in: for example, the user simply clicks on the book detail page to browse the content of the directory, finds that he is not interested and then leaves; or clicks into the reading interface of a book because of a mistake. Etc. Therefore, in order to reflect the e-book with the user's actual consumption demand, the embodiment only focuses on the e-book with deep interaction relationship, converts the user's interaction behavior with respect to the electronic into the interaction duration, and reflects the user according to the size of the interaction duration. The depth of interaction between e-books enables the screening of e-books with potential consumer needs.
  • the method is specifically implemented as follows: for each adjacent two interaction operations, respectively, according to the time interval between the adjacent two interaction operations and the preset time axis fragmentation At least one of the accumulating rules determines whether the time interval between the two adjacent interaction operations is recorded as an interaction interval; and the interaction duration of the reading user for the e-book is determined according to the recorded respective interaction intervals. It can be seen that the interaction duration in this embodiment is determined according to the recorded interaction intervals, and the determination manner of the interaction interval may be at least one of the following two manners:
  • the time interval is recorded as an interactive interval. Specifically, it is determined whether the time interval between the two adjacent interaction operations is greater than a preset interval threshold; if not, the time interval between the two adjacent interaction operations is recorded as an interaction interval. In this manner, the time interval between two adjacent interaction operations with a short interval can be recorded as an interaction interval, so that the interaction duration is counted according to this, and the two adjacent interaction operations with a long interval are not used as statistical interactions.
  • the purpose of setting the interval threshold is to constrain the concentration of multiple interactions, thereby reflecting the user's completion of the reading process, and there may be no operational correlation between the interactions with too long intervals.
  • the time interval cannot reflect the actual user operation process. For example, suppose the user puts the mobile phone on the side after opening an e-book through the e-book APP, and returns to the e-book page after one morning of work, which should obviously not be counted. The duration of the interaction.
  • each interaction operation is recorded in advance according to an operation time stamp to an interaction time axis corresponding to the e-book; and then, the interaction time axis is divided into multiple times according to a preset time axis fragmentation accumulation rule. Fragment; Finally, for each time slice, the time interval between each adjacent two interaction operations in each interaction operation belonging to the time slice is recorded as an interaction interval.
  • the purpose of setting time slicing in this manner is to constrain the degree of concentration of the interaction operation (specifically in the time interval of the interaction operation).
  • the details of the implementation of this step in the second determination mode are described in detail below, including:
  • Step 1 Record each interaction operation in advance according to the operation time stamp to the interaction time axis corresponding to the e-book.
  • the operation record of the interaction operation corresponding to each e-book is stored in a set corresponding to the e-book; then, the operation records in the set are sequentially time-ordered according to the order of the operation timestamps; finally, The sorted individual operation records are recorded in time stamps on the interactive time axis corresponding to the e-book.
  • Step 2 According to a preset time axis fragmentation accumulation rule, the interaction time axis is divided into multiple time slices.
  • the time axis is divided into a plurality of time slices according to at least one of a preset fragmentation duration and an operation timestamp of each interaction operation.
  • time slicing is divided only according to the preset fragmentation duration: assuming that the preset fragmentation duration is 30 minutes, correspondingly, the first interaction operation recorded on the interactive time axis (ie, The operation timestamp of the earliest one of the operation timestamps is sliced from the starting point, and the first time slice with a length of 30 minutes is divided; the end point of the previous previous time slice is successively the starting point of the next time slice. slice.
  • each time slice in the division mode is continuous, and there is no time interval between each adjacent two time slices, that is, each time subsequent to the first time slice, each subsequent time The fragmentation ends with the end point of the above one time slice as the starting point.
  • the time slice is divided according to the preset fragmentation duration and the operation timestamp of each interaction operation: assuming that the preset fragmentation duration is 30 minutes, correspondingly, recording on the interactive time axis
  • the operation timestamp of the first interaction (that is, the earliest one of the operation timestamps) is sliced from the starting point, and the first time slice with a length of 30 minutes is divided; after the end point of the previous previous time slice An interworking is sliced as the starting point for the next time slice.
  • each time slice in the division manner is not necessarily continuous, and there may be a time interval between two adjacent time slices.
  • the two types of fragmentation methods described above can be flexibly selected by those skilled in the art.
  • the fragmentation mode of the first mode is easier to operate, and the fragmentation mode of the second mode can reduce the number of slices and reduce the amount of calculation.
  • Step 3 For each time slice, the time interval between each adjacent two interaction operations in each interaction operation belonging to the time slice is recorded as an interaction interval.
  • the interaction durations in each time slice are respectively calculated in units of time slices. For example, suppose there are three interoperations in the first time slice, and the corresponding timestamps are 3:00, 3:05, 3:15, then the time interval between 3:00 and 3:05.
  • the 5 minute record is an interaction interval, and the interval between 3:05 and 3:15 is recorded as another interval of 10 minutes, that is, the first time interval is recorded with two interaction intervals of 5 minutes and 10 minutes.
  • the way the interaction interval in other time slices is determined is similar.
  • the advantage of this method is that it regulates the concentration of operational behavior and eliminates the statistics of the duration between the operational actions that are far away. That is, the interaction duration is only calculated according to the interaction operation within each time slice.
  • the interaction duration is not counted because it belongs to different time slices, so that the user can be more truly reflected.
  • the depth of interaction with the eBook APP usually, multiple operations generated in a short period of time (for example, 30 minutes) are more likely to be the user's operational behavior during a continuous use (interaction) process, while larger time intervals lose the true reflection of the continuous use process. .
  • the user has entered the details page, view the catalog, read the free chapter, enter the paid page, and not pay and exit the paid page in 30 minutes.
  • the sum of these operations in time series can reflect that the user continues for a period of time. Interacting with this book, the duration of this is the length of statistical interaction.
  • Step 4 For each time slice, determine the duration of the slice interaction sub-time of the time slice according to each interaction interval recorded in the time slice, and determine the read user according to the slice interaction time length of each time slice. The duration of the e-book interaction.
  • each slice calculates the length of time between all adjacent interaction operations contained therein, and sum and obtain the slice interaction time duration corresponding to the time slice (the time stamp of the operation behavior and The coincidence of the time endpoints of the fragments is also included. Then, the slice interaction time durations of all the slices are summed to obtain the total interaction duration of the reading user for the e-book.
  • the operation with too long interval can be eliminated, thereby excluding the time for the user to work or learn after opening the APP.
  • the inventor found in the process of implementing the present disclosure that the user may have some long-lasting operations in the process of reading the e-book, and the start time point and the end time point of the type operation are respectively used as an interaction operation. It is recorded on the interactive timeline.
  • the start time point and the end time point may be located in different time slices, which may cause the duration of the operation to be counted.
  • the preset persistent class is further determined. Whether the operation start time point and the operation end time point belong to the time slice; if not, calculate the operation duration according to the operation start time point and the operation end time point; according to the operation duration duration correction and the continuous class operation Corresponding at least two time sliced slice interaction sub-periods.
  • the duration of the operation needs to be counted in the final interaction duration, and the duration of the continuous interaction cannot be omitted, and the duration of the continuous interaction duration and other interactions included therein are prevented from being repeatedly counted.
  • the preset continuous class operations include: reading class operations, writing comment class operations, and the like.
  • the reading class operation behavior has a concept of duration, and the start time and end time of reading are sorted as an interaction operation on the time axis.
  • the reading duration exceeds the range of the slice, the duration of the excess portion needs to be counted in the interaction duration of the slice, and in addition, when the reading start time and the end time fall into different slices.
  • it is processed according to certain rules to prevent repeated recording duration in two fragments: for example, in one rule, each time slice is only for the reading start or end point.
  • the duration of the continuous reading behavior falling into the segment is separately counted; for example, in another rule, each time slice is separately counted for the duration of the reading start point and the duration of the continuous reading behavior of the ending point falling into the segment.
  • each time slice is separately counted for the duration of the reading start point and the duration of the continuous reading behavior of the ending point falling into the segment.
  • the operation duration corresponding to the preset continuous class operation may be further combined. Determine how long the reading user interacts with the eBook. For example, if the time interval between the operation start timestamp and the operation end timestamp of the read operation is greater than the preset value mentioned above, then the adjacent two interaction operations in each interaction operation only include the time. When the interval between the two interactions is not greater than the preset value, the duration of the operation of the reading operation cannot be taken into consideration. For this reason, the present disclosure further determines the reading user according to the duration of the operation corresponding to the preset continuous class operation. The duration of the interaction for the e-book.
  • the duration of the operation corresponding to the preset continuous class operation is further added to the interaction duration of the corresponding e-book, thereby Make the resulting interaction time more reasonable.
  • Step S240 Filter the deep interactive e-book corresponding to the reading user from the obtained e-book according to the interaction duration of the read e-book.
  • all the e-books are sorted according to the interaction duration of each e-book, and the first e-books are filtered as the deep interactive e-book according to the sorting result.
  • the obtained e-books may be filtered according to preset screening rules, so that the finally obtained deep interactive e-books are more in line with user requirements.
  • the screening rule may be determined according to factors such as the interaction duration, the operation type, and the like, and may specifically include at least one of the following rules:
  • An e-book in which the interaction duration is greater than the first preset duration and the search operation is performed. For example, an e-book with an interaction duration greater than 600 s and a user performing a search operation on the book is selected.
  • the duration of the screening interaction is greater than 1800s, and the progress of the downloaded chapters accounts for more than 80% of the entire book chapter, and the e-books with chapters greater than 200 chapters and more than 5 downloads are downloaded.
  • the e-book with the interaction duration is greater than the third preset duration and the payment behavior exists.
  • the interaction duration is greater than 3600s.
  • the rule is a rule of a single duration dimension, so the fourth preset duration should be set to a larger value.
  • Step S250 Determine the consumption demand information of the reading user according to the collected depth interactive e-book and the consumption amount of the corresponding deep interactive e-book.
  • the amount of consumption of the deep interactive e-book mainly refers to the amount of the unpaid portion corresponding to the e-book.
  • the entire purchase of an e-book refers to the entire price of the e-book; for an e-book purchased in accordance with the chapter, the price of the unpurchased chapter in the e-book.
  • the consumption demand information of the reading user may be determined according to the sum of the consumption amounts of the respective deep interactive e-books.
  • the consumer demand information poPy can be calculated by the following formula:
  • bid means book ID
  • CH bid is a book list of e-books that meet the preset rules (that is, have deep interactive relationship) (ie, a list of deep interactive e-books obtained through the above steps)
  • Price bid is in the list of books mentioned above.
  • the price of each book that is, the amount of consumption (exactly, the unpaid part price, or the potential consumption space.
  • the price needs to be calculated in conjunction with the user's chapter purchase data, that is, the remaining unpurchased chapters are calculated. Total price).
  • the book's consumption inventory demand the sum of the consumption inventory requirements of all deep interactive e-books, is the total consumption inventory demand corresponding to the reading user, that is, the consumer demand information poPy.
  • the consumer demand information in order to make the finalized consumer demand information more reasonable, it is also possible to further combine at least one piece of information to correct the consumer demand information: the historical daily consumption of the reading user, the average daily consumption of the entire network user, and the number of days of demand conversion. And the default one-day maximum recharge amount.
  • the basis for calculating the consumption demand information poPy is a book of deep interaction.
  • the size of the consumer demand information is determined by the breadth of the user's interest in the book (number of books) and depth (interaction depth).
  • the inventor found in the process of implementing the present disclosure that the user's reading interest does not necessarily match the user's consumption habits. According to statistics, the vast majority of such mismatches are determined by interest greater than the consumption habits.
  • the amount of consumption that is, the user is willing to read a lot of books, but is not willing to spend the corresponding money (poPy).
  • the poPy calculated based on the user's interest is not the actual consumer demand of the user and needs to be corrected. For example, it can be handled according to the following formula:
  • B is the corrected consumption demand information
  • mxPyD1 is the maximum value of the daily consumption amount in the history of the reading user.
  • the significance of the formula is that if the poPy is greater than the maximum value of the daily consumption amount in the history of the reading user, the non-conformity is indicated.
  • the user's consumption habits, therefore, the B value is based on mxPyD1, or combined with mxPyD1 to correct the B value, so that the final consumer demand information is more suitable for the user.
  • 3 is an adjustment coefficient for the number of days, and by setting the size of the adjustment coefficient, the tolerance of the time required for the conversion of the demand inventory can be adjusted.
  • other values such as 2 days or 5 days can also be set.
  • 3.4 is the average daily consumption quota of the users of the whole network.
  • the value needs to be adjusted according to the average daily consumption quota of the users of the whole network obtained in the business scenario.
  • the purpose of taking a larger value between this value and mxPyD1*3 is to adjust the B value to the user's average amount of the whole network when the user's single-day consumption quota is less than the average user's average amount, which can guide the user to increase the consumption quota to a certain extent.
  • the formula may also increase the coefficient of the maximum recharge amount allowed in a single day, and the modified formula is:
  • 100 is the maximum recharge amount allowed in the single day specified in the APP, and the specific value can be flexibly adjusted according to the business scenario. That is, the accumulated recharge amount of a user within one day cannot exceed 100 yuan.
  • the purpose of increasing the coefficient is that when the poPy is greater than 100, the user can only recharge 100 yuan on the same day. Therefore, the poPy greater than 100 has little significance for the subsequent operation and distribution. Therefore, the operation can be finally obtained by taking the minimum value. Consumer demand information is more closely matched to the APP.
  • the operation capable of expressing the duration of the user interaction only involves continuous operations such as reading.
  • the user can calculate the user once by using various reading duration statistics.
  • the duration of the reading behavior is that the remaining interactions are operations at the time point and do not involve the concept of length. Therefore, in this embodiment, the method of slicing is used to convert the instantaneous operation into a time length reflecting the user interaction process, and the interval is eliminated. The long-term operation of the moment improves the accuracy of the interaction time.
  • the reading user is determined according to the time interval between two adjacent interaction operations in each interaction operation performed by the reading user for each e-book.
  • the interaction time of each e-book is selected, and the deep interactive e-book corresponding to the reading user is selected according to the interaction duration, and finally the consumption demand is calculated according to the deep interactive e-book.
  • the method can select an e-book with deep interaction depth, that is, an e-book that the user is more interested in, from the plurality of e-books that interact with the user, thereby effectively improving the accuracy of the consumer demand, thereby improving the conversion rate of the active content pushed according to the user. And successfully achieve the effect of improving user reading.
  • not only the time interval of the interaction operation but also the special case of the continuous class operation are considered, so that the finally calculated consumption demand information is more accurate.
  • a third embodiment of the present application provides a non-transitory computer readable storage medium storing at least one executable instruction executable in any of the above method embodiments. The method of determining the e-book consumer demand.
  • the executable instructions may specifically be used to cause the processor to perform the following operations:
  • the consumption demand information of the reading user is determined according to the selected depth interactive e-book and the consumption amount of the corresponding deep interactive e-book.
  • executable instructions further cause the processor to:
  • each of the interaction operations includes a preset continuous class operation
  • the duration of the operation of the reading user for the e-book is determined in conjunction with the duration of the operation corresponding to the preset continuous class operation.
  • executable instructions further cause the processor to:
  • the length of interaction of the reading user for the e-book is determined based on the recorded individual interaction intervals.
  • executable instructions further cause the processor to:
  • executable instructions further cause the processor to:
  • the time interval between each adjacent two interaction operations in each interaction operation belonging to the time slice is recorded as an interaction interval
  • the step of determining the interaction duration of the reading user for the e-book according to each recorded interaction interval includes:
  • executable instructions further cause the processor to:
  • the time axis is divided into a plurality of time slices according to at least one of a preset fragmentation duration and an operation timestamp of each interaction operation.
  • executable instructions further cause the processor to:
  • each of the interaction operations belonging to the time slice includes a preset continuous class operation
  • the operation duration is calculated according to the operation start time point and the operation end time point;
  • the slice interaction sub-time length of at least two time slices corresponding to the continuous class operation is corrected according to the operation duration.
  • executable instructions further cause the processor to:
  • the e-books included in the reading history record of the reading user are culled;
  • the e-book remaining after performing the culling process is determined as an e-book that interacts with the reading user;
  • the preset culling rule includes at least one of the following: culling an e-book that has not been downloaded by the user, excluding an e-book that has not interacted within a preset period of time, and culling an e-book with a read ratio less than a preset culling ratio Book, remove free e-books, and eliminate e-books that users have paid for.
  • executable instructions further cause the processor to:
  • the deep interactive e-book corresponding to the reading user is filtered from the obtained e-book;
  • the preset screening rule includes at least one of the following:
  • An e-book with an interaction duration greater than a third preset duration and having a payment behavior is an e-book with an interaction duration greater than a third preset duration and having a payment behavior
  • An e-book with an interaction duration greater than a fourth preset duration is filtered.
  • executable instructions further cause the processor to:
  • consumer demand information by combining at least one of the following information:
  • FIG. 3 is a schematic structural diagram of an electronic device according to Embodiment 4 of the present disclosure, and the specific embodiment of the present disclosure does not limit the specific implementation of the electronic device.
  • the electronic device can be a server.
  • the electronic device can include a processor 302, a communications interface 304, a memory 306, and a communications bus 308.
  • the processor 302, the communication interface 304, and the memory 306 complete communication with each other via the communication bus 308.
  • the communication interface 304 is configured to communicate with network elements of other devices, such as clients or other servers.
  • the processor 302 is configured to execute the program 310, and specifically, the related steps in the foregoing method for determining the e-book consumption requirement.
  • program 310 can include program code, the program code including computer operating instructions.
  • the processor 302 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present disclosure.
  • the one or more processors included in the electronic device may be the same type of processor, such as one or more CPUs; or may be different types of processors, such as one or more CPUs and one or more ASICs.
  • the memory 306 is configured to store the program 310.
  • Memory 306 may include high speed RAM memory and may also include non-volatile memory, such as at least one disk memory.
  • the program 310 can be specifically configured to cause the processor 302 to perform the following operations:
  • the consumption demand information of the reading user is determined according to the selected depth interactive e-book and the consumption amount of the corresponding deep interactive e-book.
  • executable instructions cause processor 302 to perform the following operations:
  • each of the interaction operations includes a preset continuous class operation
  • the duration of the operation of the reading user for the e-book is determined in conjunction with the duration of the operation corresponding to the preset continuous class operation.
  • the executable instructions cause processor 302 to perform the following operations:
  • the length of interaction of the reading user for the e-book is determined based on the recorded individual interaction intervals.
  • executable instructions cause processor 302 to perform the following operations:
  • the executable instructions cause processor 302 to perform the following operations:
  • the time interval between each adjacent two interaction operations in each interaction operation belonging to the time slice is recorded as an interaction interval
  • the steps of determining the interaction duration of the reading user for the e-book according to the recorded interaction intervals include:
  • executable instructions cause processor 302 to perform the following operations:
  • the time axis is divided into a plurality of time slices according to at least one of a preset fragmentation duration and an operation timestamp of each interaction operation.
  • the executable instructions cause processor 302 to perform the following operations:
  • each of the interaction operations belonging to the time slice includes a preset continuous class operation
  • the operation duration is calculated according to the operation start time point and the operation end time point;
  • the slice interaction sub-time length of at least two time slices corresponding to the continuous class operation is corrected according to the operation duration.
  • executable instructions cause processor 302 to perform the following operations:
  • the e-books included in the reading history record of the reading user are culled;
  • the e-book remaining after performing the culling process is determined as an e-book that interacts with the reading user;
  • the preset culling rule includes at least one of the following: culling an e-book that has not been downloaded by the user, excluding an e-book that has not interacted within a preset period of time, and culling an e-book with a read ratio less than a preset culling ratio Book, remove free e-books, and eliminate e-books that users have paid for.
  • executable instructions cause processor 302 to perform the following operations:
  • the deep interactive e-book corresponding to the reading user is filtered from the obtained e-book;
  • the preset screening rule includes at least one of the following:
  • An e-book with an interaction duration greater than a third preset duration and having a payment behavior is an e-book with an interaction duration greater than a third preset duration and having a payment behavior
  • An e-book with an interaction duration greater than a fourth preset duration is filtered.
  • executable instructions cause processor 302 to perform the following operations:
  • consumer demand information by combining at least one of the following information:
  • modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment.
  • the modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components.
  • any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods or devices so disclosed All processes or units are combined.
  • Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • User Interface Of Digital Computer (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种电子书消费需求的确定方法、电子设备及计算机存储介质，该方法包括：获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳（S110）；根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长（S120）；根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书（S120）；根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息（S140）。

Description

电子书消费需求的确定方法、电子设备及计算机存储介质
相关申请的交叉参考
本申请要求于2018年3月30日提交中国专利局、申请号为201810299147.3、名称为“电子书消费需求的确定方法、电子设备及计算机存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本公开涉及计算机领域，具体涉及一种电子书消费需求的确定方法、电子设备及计算机存储介质。
背景技术
随着用户阅读意识的提升，电子书得到了越来越广泛的普及。为了更好地向阅读用户推送与电子书相关的活动信息，往往需要结合阅读用户的消费需求来确定与当前的阅读用户相匹配的活动种类，从而提升推送成功率，进而提升用户的阅读量。目前，在确定阅读用户的消费需求时，一般根据阅读用户的历史阅读记录将该阅读用户点击、下载或阅读过的电子书全部加入需求列表中，并根据该需求列表计算该阅读用户的消费需求。
但是，在实现本公开的过程中，发明人发现：简单地将阅读用户点击、下载或阅读过的电子书全部加入需求列表的方式过于简单粗放，并未考虑各个电子书与阅读用户之间的交互深度，容易将用户因误操作而点击的电子书，或用户完全不感兴趣、仅浏览数秒就关闭的电子书加入需求列表中，从而导致最终得到的消费需求无法真实反映用户的潜在消费能力，进而导致据此推送的活动内容转化率不高，无法实现提升用户阅读量的效果。
发明内容
鉴于上述问题，提出了本公开以便提供一种克服上述问题或者至少部分地解决上述问题的电子书消费需求的确定方法、电子设备及计算机存储介质。
根据本公开的一个方面，提供了一种电子书消费需求的确定方法，包括：
获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳；
根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长；
根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的 消费需求信息。
根据本公开的另一个方面，提供了一种电子设备，包括：处理器、存储器、通信接口和通信总线，处理器、存储器和通信接口通过通信总线完成相互间的通信；存储器用于存放至少一可执行指令，可执行指令使处理器执行以下操作：
获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳；
根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长；
根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息。
根据本公开的又一方面，提供了一种非易失性计算机可读存储介质，该非易失性计算机可读存储介质中存储有至少一可执行指令，可执行指令使处理器执行以下操作：
获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳；
根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长；
根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息。
根据本公开的再一方面，还提供了一种计算机程序产品，该计算机程序产品包括存储在上述非易失性计算机可读存储介质上的计算程序。
在本公开提供的电子书消费需求的确定方法、电子设备及计算机存储介质中，根据阅读用户针对每本电子书执行的各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对每本电子书的交互时长，并根据交互时长筛选与阅读用户相对应的深度交互电子书，最终根据深度交互电子书计算消费需求。与现有技术中将用户操作过的电子书确定为需求电子书相比，本公开以用户针对电子书产生的一系列交互操作为依据、以交互操作之间的时间间隔作为考量维度，量化用户与电子书的交互深度，并以此反映用户阅读该电子书的渴望程度，同时进一步结合电子书的消费金额，将用户对电子书的渴望程度转换为用户对该本书的潜在消费需求，由此提高运营数据的价值，提升运营活动的转化率。
上述说明仅是本公开技术方案的概述，为了能够更清楚了解本公开的技术手段，而可依照说明书的内容予以实施，并且为了让本公开的上述和其它目的、特征和优点能够更明 显易懂，以下特举本公开的具体实施方式。
附图概述
通过阅读下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本公开的限制。而且在整个附图中，用相同的参考符号表示相同的部件。在附图中：
图1示出了本公开实施例一提供的电子书消费需求的确定方法的流程图；
图2示出了本公开实施例二提供的电子书消费需求的确定方法的流程图；
图3示出了根据本公开实施例四的一种电子设备的结构示意图。
本公开的较佳实施方式
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例，然而应当理解，可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反，提供这些实施例是为了能够更透彻地理解本公开，并且能够将本公开的范围完整的传达给本领域的技术人员。
实施例一
图1示出了本公开实施例一提供的电子书消费需求的确定方法的流程图。如图1所示，该方法包括以下步骤：
步骤S110：获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳。
首先，确定与阅读用户发生交互的电子书时，可以将阅读用户的历史阅读记录中所包含的每一本电子书均作为与阅读用户发生交互的电子书。其中，历史阅读记录用于记录阅读用户的各类阅读行为及其对应的电子书信息。其中，阅读行为包括：下载、点击、阅读、打开、付费(这里是指对于连载书籍的前部分章节的付费行为，而非完整付费行为)等可能发生的各种行为。该方式能够避免遗漏，确保收集到尽可能全面的信息。
或者，也可以通过下述方式确定与阅读用户发生交互的电子书：针对阅读用户的历史阅读记录中所包含的每一本电子书，判断该阅读用户是否针对该电子书执行过预设交互操作，若是，则将其确定为与阅读用户发生交互的电子书。其中，预设交互操作可以仅包括特定类型的、能够反映出用户对电子书具备初步兴趣的操作，例如，搜索操作、下载操作、进入书籍详情页操作、查看目录操作等。该方式能够直接获取到用户可能比较感兴趣的电子书，以提升后续的处理效率。
接下来，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳。具体地，对于每一本电子书，当阅读用户针对该电子书执行过多次交互操作时，分别获取每一次交互操作的操作时间戳。其中，操作时间戳是指：执行该次交互操 作时所对应的系统时间。由此可见，通过各次交互操作的操作时间戳能够确定各次交互操作的操作顺序。
步骤S120：根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长。
具体地，在根据操作时间戳确定出各次交互操作的操作顺序之后，进一步确定各次交互操作中相邻的两次交互操作之间的时间间隔。其中，在一种实现方式中，各次交互操作中相邻的两次交互操作包括各次交互操作中每相邻的两次交互操作，即：将每相邻的两次交互操作均考虑在内。在又一种实现方式中，各次交互操作中相邻的两次交互操作仅包括符合预设规则的若干组相邻的两次交互操作。例如，仅包括时间间隔不大于预设数值的相邻两次交互操作，从而滤除时间间隔过长的交互操作。又如，仅包括预设时段之内(例如最近三个月)的相邻两次交互操作，从而排除时间过久的交互操作。本公开不限定预设规则的具体内容和实现方式，只要能够更加真实地反映用户需求即可。
在根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长时，可以对各次交互操作中相邻的两次交互操作之间的时间间隔进行预设处理，例如累加求和等，从而得到阅读用户针对该电子书的交互时长。本公开对预设处理的具体方式不做限定。
另外，发明人在实现本公开的过程中发现：在阅读过程中除了瞬时完成的交互操作之外，还可能包含阅读、写评论等持续时间较长的操作，该类操作也能够反映用户与电子书的真实交互过程，因此，为了更加准确地反映用户与电子书之间的交互程度，可选地，在本步骤中，当确定各次交互操作中包含预设持续类操作时，还可以进一步结合预设持续类操作所对应的操作持续时长，确定阅读用户针对该电子书的交互时长。其中，持续类操作是指：相对进入阅读页、打开目录等瞬时完成的操作而言，操作过程需要持续一段时间完成的操作，例如，阅读操作、写评论操作等各种持续时间较长的操作。
步骤S130：根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书。
由于交互时长是根据阅读用户针对电子书的交互操作得到的，能够反映阅读用户与该电子书之间的交互深度，因此，根据交互时长能够筛选出与阅读用户相对应的深度交互电子书。
例如，在一种实现方式中，可以按照各本电子书的交互时长对全部电子书进行排序，根据排序结果筛选前若干本电子书作为深度交互电子书。又如，在又一种实现方式中，可以按照预设筛选规则筛选与阅读用户相对应的深度交互电子书。其中，筛选规则可根据交互时长、操作类型等因素确定。上述两种方式既可以单独使用，也可以结合使用。
需要说明的是，本实施例中所述的深度交互电子书是指，在对可量化指标数据的衡量过程中，指标数据满足预定阈值的电子书。例如当可量化指标数据为交互时长时，预定阈 值可以是根据交互时长排序后的前序数量TOP N(正整数)，排序在前N本的电子书为深度交互电子书；或者，预定阈值可以是预设的交互时长，大于等于该交互时长的电子书为深度交互电子书；或者，当可量化指标数据为特定操作类型产生的次数时，预定阈值可以是具体设定的次数M(正整数)，例如当用户先后2次进入某本电子书的购买界面(但是均未购买)时，表明用户对该电子书有一定的阅读兴趣但对付费购买犹豫再三，因此可将该电子书认定为深度交互电子书。
步骤S140：根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息。
其中，深度交互电子书的消费金额主要是指：该电子书所对应的未付费部分的金额。例如，对于整本购买的电子书，是指该电子书的整本价格；对于按章购买的电子书，是指该电子书中的未购买章节的价格。具体实施时，可以根据各本深度交互电子书的消费金额总和来确定阅读用户的消费需求信息。
由此可见，在本公开提供的电子书消费需求的确定方法中，根据阅读用户针对每本电子书执行的各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对每本电子书的交互时长，并根据交互时长筛选与阅读用户相对应的深度交互电子书，最终根据深度交互电子书计算消费需求。与现有技术中将用户操作过的电子书确定为需求电子书相比，本公开以用户针对电子书产生的一系列交互操作为依据、以交互操作之间的时间间隔作为考量维度，量化用户与电子书的交互深度，并以此反映用户阅读该电子书的渴望程度，同时进一步结合电子书的消费金额，将用户对电子书的渴望程度转换为用户对该本书的潜在消费需求，由此提高运营数据的价值，提升运营活动的转化率。
实施例二
图2示出了本公开实施例二提供的电子书消费需求的确定方法的流程图。如图2所示，该方法包括以下步骤：
步骤S210：获取与阅读用户发生交互的电子书。
具体地，本步骤可通过下述方式实现：确定阅读用户的历史阅读记录中包含的各个电子书；按照预设剔除规则，对阅读用户的历史阅读记录中包含的各个电子书进行剔除处理；将执行剔除处理之后剩余的电子书确定为与阅读用户发生交互的电子书。
具体实施时，在确定阅读用户的历史阅读记录中包含的各个电子书时，既可以查找阅读用户的历史阅读记录中包含的每一本电子书，也可以仅查找阅读用户执行过预设交互操作的电子书，具体实现细节可参照步骤S110中相应部分的描述，此处不再赘述。接下来，按照预设剔除规则，对查找到的各个电子书进行剔除处理。其中，预设剔除规则用于剔除与消费需求无关的电子书，即：对于计算消费需求没有贡献的电子书，具体包括：用户不感兴趣(即交互深度不够)的电子书和无潜在消费空间的电子书(例如免费的或已完整付费的电子书)中的至少一个。在本实施例中，预设剔除规则可包括以下中的至少一个：
(1)剔除用户未下载过的电子书：若用户未下载过该电子书，则可以推测用户对该电子书不感兴趣，应当剔除。
(2)剔除预设时段内未发生交互的电子书：例如，若近一个月之内，用户与该电子书未发生任何交互行为，则说明用户已将该书遗忘。
(3)剔除免费试读章节的已读比例小于预设剔除比例的电子书：例如，用户对免费试读章节部分的阅读进度不足预定比例，则说明用户对该书不感兴趣。
(4)剔除免费的电子书、以及剔除用户已完整付费(包括整本购买和按章购买)的电子书：该类书籍为已完整付费书籍，因而没有潜在消费空间，即：该类书籍所对应的消费金额为零，为了节省后续的计算量，可在本步骤中直接剔除。
步骤S220：针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳。
具体地，历史阅读记录中包含阅读用户针对各个电子书执行的各次交互操作的操作记录，其中，每个操作记录中包含执行该次交互操作时的操作时间戳以及该次交互操作的操作对象(即电子书)。相应地，通过分析各个操作记录，能够针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳。
步骤S230：根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长。
在步骤S210中获取到的电子书中可能包含用户不感兴趣的电子书：例如用户只是点进书籍详情页浏览一下目录内容，发现不感兴趣然后离开；或者因为误操作点击进入某本书的阅读界面等，因此，为反映出用户具有切实消费需求的电子书，本实施例仅关注具有深度交互关系的电子书，将用户针对电子的交互行为转化为交互时长，并根据交互时长的大小体现用户与电子书之间的交互深度，从而实现对具有潜在消费需求的电子书的筛选。
为了达到上述目的，在本实施例中，具体通过如下方式实现：分别针对每相邻的两次交互操作，根据该相邻的两次交互操作之间的时间间隔和预设的时间轴分片累加规则中的至少一个，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔；根据已记录的各个交互间隔确定阅读用户针对该电子书的交互时长。由此可见，本实施例中的交互时长根据已记录的各个交互间隔确定，其中，交互间隔的确定方式可采用下述两种方式中的至少一种：
在第一种交互间隔的确定方式中，分别针对每相邻的两次交互操作，根据该相邻的两次交互操作之间的时间间隔，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。具体地，判断该相邻的两次交互操作之间的时间间隔是否大于预设间隔阈值；若否，将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。该种方式能够将间隔较短的两次相邻交互操作之间的时间间隔记录为一个交互间隔，以便据此统计交互时 长，而对于间隔较长的两次相邻交互操作则不作为统计交互时长的依据，设定间隔阈值的目的在于，约束多次交互操作的集中程度，由此反映用户的一次完成阅读过程，而间隔时间过长的交互操作之间可能并不存在操作上的关联性，其时间间隔无法反映真实的用户操作过程，例如，假设用户通过电子书APP打开一本电子书之后将手机搁置一边，工作一上午之后再回到电子书页面，该段时间明显不应计入交互时长。
在第二种交互间隔的确定方式中，分别针对每相邻的两次交互操作，根据预设的时间轴分片累加规则，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。具体实施时，预先将各次交互操作按照操作时间戳记录到与该电子书相对应的交互时间轴上；然后，按照预设的时间轴分片累加规则，将交互时间轴划分为多个时间分片；最后，分别针对每个时间分片，将属于该时间分片的各次交互操作中每相邻的两次交互操作之间的时间间隔记录为一个交互间隔。与前述方式类似，本方式中设置时间分片目的在于约束交互操作的集中程度(具体体现在交互操作的时间间隔上)。下面详细介绍采用第二种确定方式时本步骤的具体实现细节，包括：
步骤一、预先将各次交互操作按照操作时间戳记录到与该电子书相对应的交互时间轴上。
首先，将每本电子书所对应的交互操作的操作记录存储到与该本电子书相对应的集合中；然后，按照操作时间戳的顺序对该集合中的各个操作记录进行时序排序；最后，将排序后的各个操作记录按照时间戳记录到与该电子书相对应的交互时间轴上。
步骤二、按照预设的时间轴分片累加规则，将交互时间轴划分为多个时间分片。
具体地，根据预设分片时长和各次交互操作的操作时间戳中的至少一个，将时间轴划分为多个时间分片。例如，在第一种划分方式中，仅根据预设分片时长来划分时间分片：假设预设分片时长为30分钟，相应地，以交互时间轴上记录的第一个交互操作(即操作时间戳最早的一个交互操作)的操作时间戳为起点进行切片，划分出长度为30分钟的第一时间分片；后续以前一个时间分片的结束点作为下一个时间分片的起始点连续切片。由此可见，该种划分方式中的各个时间分片为连续的，每相邻的两个时间分片之间不存在时间间隔，即：除第一个时间分片外，后续的每个时间分片均以上一个时间分片的结束点作为起始点。又如，在第二种划分方式中，结合预设分片时长以及各次交互操作的操作时间戳划分时间分片：假设预设分片时长为30分钟，相应地，以交互时间轴上记录的第一个交互操作(即操作时间戳最早的一个交互操作)的操作时间戳为起点进行切片，划分出长度为30分钟的第一时间切片；后续以前一个时间分片的结束点之后的第一个交互操作作为下一个时间分片的起始点进行切片。由此可见，该种划分方式中的各个时间分片不一定是连续的，相邻的两个时间分片之间有可能存在时间间隔。上述的两种分片划分方式本领域技术人员可以灵活选择，第一种方式的分片方式更易于操作，第二种方式的分片方式能够减少分片数量、降低计算量。
步骤三、分别针对每个时间分片，将属于该时间分片的各次交互操作中每相邻的两次交互操作之间的时间间隔记录为一个交互间隔。
具体地，本步骤分别以各个时间分片为单位计算各个时间分片内部的交互时长。例如，假设在第一个时间分片内具有三个交互操作，其对应的时间戳分别为3:00、3:05、3:15，则将3:00与3:05之间的时间间隔5分钟记录为一个交互间隔，将3:05与3:15之间的时间间隔10分钟记录为又一个交互间隔，即：第一个时间分片内记录有两个交互间隔5分钟和10分钟，其他时间分片内的交互间隔的确定方式与之类似。该种方式的优势在于：对操作行为的集中度进行规制，剔除距离较远的操作行为之间的时长统计。即：仅根据各个时间分片内部的交互操作统计交互时长，对于间隔时间很长的两次交互操作，由于其分别属于不同的时间分片则不予统计交互时长，从而能够更加真实地反映用户与电子书APP之间的交互深度。通常，在短时间内(例如30分钟)产生的多个操作，更可能是用户在一段持续使用(交互)过程中产生的操作行为，而较大的时间间隔则失去对持续使用过程的真实反映。例如用户在30分钟内先后进行了进入详情页、查看目录、阅读免费章节、进入付费页面、未付费并退出付费页面等操作，这些操作在时序上的总和能够反映出用户在一段时间内持续的与这本书籍在进行交互，这个持续的时间长短即为统计的交互时长。反之，如果用户点击详情页之后睡觉或工作学习，几小时之后再回来打开APP做后续操作，那么这几个小时的时间显然不是用户与书籍交互的时间，需要剔除掉，本实施例通过切片的方式，通过分片长度控制剔除时长过大不合常理的情况。
步骤四、分别针对每个时间分片，根据该时间分片内记录的各个交互间隔确定该时间分片的分片交互子时长，根据各个时间分片的分片交互子时长确定阅读用户针对该电子书的交互时长。
具体地，对于每个分片，计算其中包含的、所有相邻交互操作之间的时间长度，并对此求和，获得对应该时间分片的分片交互子时长(操作行为的时间戳与分片的时间端点重合的情况也算包含在内)。然后，将所有分片的分片交互子时长进行加总，获得阅读用户针对该电子书的总交互时长。
通过上述方式能够剔除间隔时间过久的操作，从而将用户打开APP后工作或学习的时间排除在外。但是，发明人在实现本公开的过程中发现，用户在阅读电子书的过程中可能会出现一些持续时间很长的操作，该类操作的起始时间点与结束时间点将分别作为一个交互操作记录到交互时间轴上，但是，由于持续时间过长，可能导致起始时间点与结束时间点分别位于不同的时间分片中，进而导致该类操作的持续时间无法统计在内。为了解决上述问题，可选地，在上述步骤中，分别针对每个时间分片，当确定属于该时间分片的各次交互操作中包含预设持续类操作时，进一步判断该预设持续类操作的操作起始时间点与操作结束时间点是否同属于该时间分片；若否，根据操作起始时间点以及操作结束时间点计算操作持续时长；根据操作持续时长修正与该持续类操作相对应的至少两个时间分片的 分片交互子时长。具体修正时，需要将该操作持续时长统计到最终的交互时长中，既不能遗漏该持续交互时长，又要防止该持续交互时长与其中包含的其他交互操作重复统计的情况。其中，预设持续类操作包括：阅读类操作、写评论类操作等。例如，阅读类操作行为具有时长的概念，阅读的起始时间和结束时间都要作为交互操作排序到时间轴上。在进行切片后，阅读时长若超出分片的范围，则需要将超出部分的时长统计到该分片的交互时长中，另外，对于阅读起始时间和结束时间分别落入不同分片的情况，在分别统计分片交互时长时，需按照一定的规则进行处理，以防止在两个分片中重复记录时长：例如，在一种规则中，每个时间分片仅针对阅读起始点或结束点落入该分片的持续阅读行为的时长另行统计；又如，在另一种规则中，每个时间分片同时针对阅读起始点以及结束点落入该分片的持续阅读行为的时长另行统计，但统计之后要注意对重复统计的部分进行去重处理。
同理，在第一种交互间隔的确定方式中，为了提升准确性，当确定各次交互操作中包含预设持续类操作时，也可以进一步结合该预设持续类操作所对应的操作持续时长确定阅读用户针对该电子书的交互时长。举例而言，假设阅读操作的操作起始时间戳与操作结束时间戳之间的时间间隔大于上文提到的预设数值，则当各次交互操作中相邻的两次交互操作仅包括时间间隔不大于预设数值的相邻两次交互操作时，本次阅读操作的操作持续时长将无法考虑在内，为此，本公开进一步结合预设持续类操作所对应的操作持续时长确定阅读用户针对该电子书的交互时长。即：在获取时间间隔不大于预设数值的相邻两次交互操作的时间间隔的基础上，进一步将预设持续类操作所对应的操作持续时长补充计入对应电子书的交互时长中，从而使最终得到的交互时长更为合理。
步骤S240：根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书。
在本实施例中，按照各本电子书的交互时长对全部电子书进行排序，根据排序结果筛选前若干本电子书作为深度交互电子书。可选地，在筛选深度交互电子书之前或之后，还可以按照预设筛选规则对获取到的电子书进行筛选，以使最终得到的深度交互电子书更加符合用户需求。其中，筛选规则可根据交互时长、操作类型等因素确定，具体可包括以下规则中的至少一个：
(1)筛选交互时长大于第一预设时长，且执行过搜索操作的电子书。例如，筛选交互时长大于600s，并且用户针对该书籍执行过搜索操作的电子书。
(2)筛选交互时长大于第二预设时长并且具有以下特点中的至少一种的电子书：已下载章节占整书章节的比例大于预设筛选比例、已下载章节的数量大于预设筛选数量、和已下载次数大于预设次数。例如，筛选交互时长大于1800s，并且下载过的章节进度占整本书章节比例超过80％，并且下载章节大于200章，并且下载次数大于5次的电子书。
(3)筛选交互时长大于第三预设时长，且存在付费行为的电子书。例如，筛选交互时长大于1800s，并且有付费行为，并且付费金额大于0.5元的电子书。
(4)筛选交互时长大于第四预设时长的电子书。例如，交互时长大于3600s。该规则为单一时长维度的规则，因此，第四预设时长应设置为较大的数值。
步骤S250：根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息。
其中，深度交互电子书的消费金额主要是指：该电子书所对应的未付费部分的金额。例如，对于整本购买的电子书，是指该电子书的整本价格；对于按章购买的电子书，是指该电子书中的未购买章节的价格。具体实施时，可以根据各本深度交互电子书的消费金额总和来确定阅读用户的消费需求信息。
具体地，消费需求信息poPy可通过如下公式计算：
Figure PCTCN2019073492-appb-000001
其中，bid意为书籍ID，CH bid为符合预设规则(即具有深度交互关系)的电子书的书籍列表(即通过上述步骤得到的深度交互电子书的列表)，Price bid为上述书籍列表中，对应各本书的价格，即消费金额(确切说是未付部分价格，或者说潜在消费空间，对于按章购买书籍，需要结合用户的章节购买数据计算该价格，即计算剩余未购买章节的总价)。通过每本书与对应的价格相乘，得到该本书的消费存量需求，所有深度交互电子书的消费存量需求之和，即为对应该阅读用户的总消费存量需求，即消费需求信息poPy。
另外，为了使最终确定的消费需求信息更加合理，还可以进一步结合至少一项信息，对得到消费需求信息进行修正：阅读用户的历史日消费额、全网用户的平均日消费额、需求转化天数、以及预设的单日最大充值金额。
在本实施例中，计算消费需求信息poPy的基础在于深度交互的书籍，换句话说，消费需求信息的大小是由用户对书籍的兴趣的广度(书籍数量)和深度(交互深度)共同决定的。但是，发明人在实现本公开的过程中发现：用户的阅读兴趣与用户的消费习惯并不一定是匹配的，经统计发现，这种不匹配的绝大多数情况都是兴趣大于消费习惯所决定的消费额，即用户愿意看很多书，但不愿意花对应的钱(poPy)。这种情况下，根据用户兴趣计算出来的poPy并不是用户实际的消费需求，需要对其进行校正。例如，可根据下述公式处理：
B＝min(mxPyD1，poPy)
其中，B为校正后的消费需求信息，mxPyD1为该阅读用户历史上日消费额的最大值，该公式的意义在于，若poPy大于该阅读用户历史上日消费额的最大值，则说明不符合用户的消费习惯，因此，B值以mxPyD1为准，或者，结合mxPyD1对B值进行修正，以使最终得到的消费需求信息更加适合该用户。
进一步地，考虑到消费存量需求反映的是用户全部的消费需求，而这种需求量并非在一天之内能够全部转化完毕，可能需要连续消费若干天才可将需求全部转化完。相应地， 在上述公式中还可以增加一个关于天数的调整系数，变形后的公式为：
B＝min(mxPyD1*3，poPy)
其中，3为关于天数的调整系数，通过对该调整系数大小的设置，可以调整需求存量转化的时间的容忍度。具体业务场景中，除了设置为3天外，也可以设置为2天或5天等其他数值。
进一步地，对于mxPyD1*3还可做进一步修正，变形后的公式为：
B＝min(max(3.4*3，mxPyD1*3)，poPy)
其中，3.4为全网用户平均的日消费额度，实际实施时，该数值需根据业务场景中得到的全网用户平均的日消费额度进行调整。该数值与mxPyD1*3之间取较大值的目的在于：当用户单日消费额度不足全网用户平均值时，参考全网用户平均数额调整B值，一定程度上可以引导用户提高消费额度。
进一步地，公式中还可增加单日允许最大充值金额的系数，变形后的公式为：
B＝min(max(3.4*3，mxPyD1*3)，poPy，100)
其中，100为APP中规定的单日允许最大充值金额，其具体数值可根据业务场景灵活调整。即一个用户一天之内累计充值金额不能大于100元。增加该系数目的在于：当poPy大于100时，用户也只能在当日充值100元，因此，poPy大于100对后续运营分发的参考意义不大，因此，通过取最小值的操作能够使最终得到的消费需求信息与APP更加匹配。
另外，严格意义上讲，本实施例的交互操作行为中，能够表达用户交互时长长度的操作仅涉及阅读等持续类操作，在本实施例中可以使用各种阅读时长统计的手段计算出用户一次阅读行为的时长，至于剩余的交互行为都是时间点上的操作，不涉及长度的概念，因此本实施例中采用切片的方式将瞬间操作转化为反映用户交互过程的时间长度，并通过剔除间隔过久的瞬间操作的方式提升交互时长的准确性。
综上可知，在本公开提供的电子书消费需求的确定方法中，根据阅读用户针对每本电子书执行的各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对每本电子书的交互时长，并根据交互时长筛选与阅读用户相对应的深度交互电子书，最终根据深度交互电子书计算消费需求。该方式能够从与用户发生交互的众多电子书中筛选出交互深度较深，即用户较感兴趣的电子书，从而有效提升消费需求的准确度，进而提升据此推送的活动内容的转化率，并成功实现提升用户阅读量的效果。并且，在本实施例中，不仅考虑了交互操作的时间间隔，也考虑了持续类操作的特殊情况，从而使最终计算得到的消费需求信息更加准确。
实施例三
本申请实施例三提供了一种非易失性计算机可读存储介质，该非易失性计算机可读存储介质存储有至少一可执行指令，该计算机可执行指令可执行上述任意方法实施例中的电 子书消费需求的确定方法。
可执行指令具体可以用于使得处理器执行以下操作：
获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳；
根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长；
根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
当确定各次交互操作中包含预设持续类操作时，结合预设持续类操作所对应的操作持续时长确定阅读用户针对该电子书的交互时长。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
分别针对每相邻的两次交互操作，根据该相邻的两次交互操作之间的时间间隔和预设的时间轴分片累加规则中的至少一个，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
根据已记录的各个交互间隔确定阅读用户针对该电子书的交互时长。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
判断该相邻的两次交互操作之间的时间间隔是否大于预设间隔阈值；若否，将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
预先将各次交互操作按照操作时间戳记录到与该电子书相对应的交互时间轴上；
按照预设的时间轴分片累加规则，将交互时间轴划分为多个时间分片；
分别针对每个时间分片，将属于该时间分片的各次交互操作中每相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
则根据已记录的各个交互间隔确定阅读用户针对该电子书的交互时长的步骤具体包括：
分别针对每个时间分片，根据该时间分片内记录的各个交互间隔确定该时间分片的分片交互子时长，根据各个时间分片的分片交互子时长确定阅读用户针对该电子书的交互时长。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
根据预设分片时长和各次交互操作的操作时间戳中的至少一个，将时间轴划分为多个时间分片。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
分别针对每个时间分片，当确定属于该时间分片的各次交互操作中包含预设持续类操作时，进一步判断预设持续类操作的操作起始时间点与操作结束时间点是否同属于该时间分片；
若否，根据操作起始时间点以及操作结束时间点计算操作持续时长；
根据操作持续时长修正与该持续类操作相对应的至少两个时间分片的分片交互子时长。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
确定阅读用户的历史阅读记录中包含的各个电子书；
按照预设剔除规则，对阅读用户的历史阅读记录中包含的各个电子书进行剔除处理；
将执行剔除处理之后剩余的电子书确定为与阅读用户发生交互的电子书；
其中，预设剔除规则包括以下中的至少一个：剔除用户未下载过的电子书、剔除预设时段内未发生交互的电子书、剔除免费试读章节的已读比例小于预设剔除比例的电子书、剔除免费的电子书、以及剔除用户已完整付费的电子书。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
按照预设筛选规则，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
其中，预设筛选规则包括以下中的至少一个：
筛选交互时长大于第一预设时长，且执行过搜索操作的电子书；
筛选交互时长大于第二预设时长并且具有以下特征中的至少一种特征的电子书：已下载章节占整书章节的比例大于预设筛选比例、已下载章节的数量大于预设筛选数量、和已下载次数大于预设次数；
筛选交互时长大于第三预设时长，且存在付费行为的电子书；以及，
筛选交互时长大于第四预设时长的电子书。
在一种可选的方式中，其中，可执行指令进一步使处理器执行以下操作：
进一步结合下述至少一项信息，对消费需求信息进行修正：
阅读用户的历史日消费额、全网用户的平均日消费额、需求转化天数、以及预设的单日最大充值金额。
实施例四
图3示出了根据本公开实施例四的一种电子设备的结构示意图，本公开具体实施例并不对电子设备的具体实现做限定。例如，该电子设备可以为服务器。
如图3所示，该电子设备可以包括：处理器(processor)302、通信接口(Communications Interface)304、存储器(memory)306、以及通信总线308。
其中：处理器302、通信接口304、以及存储器306通过通信总线308完成相互间的通信。通信接口304，用于与其它设备比如客户端或其它服务器等的网元通信。处理器302， 用于执行程序310，具体可以执行上述电子书消费需求的确定方法实施例中的相关步骤。
具体地，程序310可以包括程序代码，该程序代码包括计算机操作指令。
处理器302可能是中央处理器CPU，或者是特定集成电路ASIC(Application Specific Integrated Circuit)，或者是被配置成实施本公开实施例的一个或多个集成电路。电子设备包括的一个或多个处理器，可以是同一类型的处理器，如一个或多个CPU；也可以是不同类型的处理器，如一个或多个CPU以及一个或多个ASIC。
存储器306，用于存放程序310。存储器306可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。
程序310具体可以用于使得处理器302执行以下操作：
获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定阅读用户针对该电子书执行的各次交互操作的操作时间戳；
根据各次交互操作中相邻的两次交互操作之间的时间间隔，确定阅读用户针对该电子书的交互时长；
根据阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
根据筛选出的深度交互电子书以及对应深度交互电子书的消费金额，确定阅读用户的消费需求信息。
在一种可选的方式中，可执行指令使处理器302执行以下操作：
当确定各次交互操作中包含预设持续类操作时，结合预设持续类操作所对应的操作持续时长确定阅读用户针对该电子书的交互时长。在一种可选的方式中，可执行指令使处理器302执行以下操作：
分别针对每相邻的两次交互操作，根据该相邻的两次交互操作之间的时间间隔和预设的时间轴分片累加规则中的至少一个，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
根据已记录的各个交互间隔确定阅读用户针对该电子书的交互时长。
在一种可选的方式中，可执行指令使处理器302执行以下操作：
判断该相邻的两次交互操作之间的时间间隔是否大于预设间隔阈值；若否，将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。在一种可选的方式中，可执行指令使处理器302执行以下操作：
预先将各次交互操作按照操作时间戳记录到与该电子书相对应的交互时间轴上；
按照预设的时间轴分片累加规则，将交互时间轴划分为多个时间分片；
分别针对每个时间分片，将属于该时间分片的各次交互操作中每相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
则根据已记录的各个交互间隔确定阅读用户针对该电子书的交互时长的步骤具体包 括：
分别针对每个时间分片，根据该时间分片内记录的各个交互间隔确定该时间分片的分片交互子时长，根据各个时间分片的分片交互子时长确定阅读用户针对该电子书的交互时长。
在一种可选的方式中，可执行指令使处理器302执行以下操作：
根据预设分片时长和各次交互操作的操作时间戳中的至少一个，将时间轴划分为多个时间分片。在一种可选的方式中，可执行指令使处理器302执行以下操作：
分别针对每个时间分片，当确定属于该时间分片的各次交互操作中包含预设持续类操作时，进一步判断预设持续类操作的操作起始时间点与操作结束时间点是否同属于该时间分片；
若否，根据操作起始时间点以及操作结束时间点计算操作持续时长；
根据操作持续时长修正与该持续类操作相对应的至少两个时间分片的分片交互子时长。
在一种可选的方式中，可执行指令使处理器302执行以下操作：
确定阅读用户的历史阅读记录中包含的各个电子书；
按照预设剔除规则，对阅读用户的历史阅读记录中包含的各个电子书进行剔除处理；
将执行剔除处理之后剩余的电子书确定为与阅读用户发生交互的电子书；
其中，预设剔除规则包括以下中的至少一个：剔除用户未下载过的电子书、剔除预设时段内未发生交互的电子书、剔除免费试读章节的已读比例小于预设剔除比例的电子书、剔除免费的电子书、以及剔除用户已完整付费的电子书。
在一种可选的方式中，可执行指令使处理器302执行以下操作：
按照预设筛选规则，从获取到的电子书中筛选与阅读用户相对应的深度交互电子书；
其中，预设筛选规则包括以下中的至少一个：
筛选交互时长大于第一预设时长，且执行过搜索操作的电子书；
筛选交互时长大于第二预设时长并且具有以下特征中的至少一种特征的电子书：已下载章节占整书章节的比例大于预设筛选比例、已下载章节的数量大于预设筛选数量、和已下载次数大于预设次数；
筛选交互时长大于第三预设时长，且存在付费行为的电子书；以及，
筛选交互时长大于第四预设时长的电子书。
在一种可选的方式中，可执行指令使处理器302执行以下操作：
进一步结合下述至少一项信息，对消费需求信息进行修正：
阅读用户的历史日消费额、全网用户的平均日消费额、需求转化天数、以及预设的单日最大充值金额。
在此提供的算法和显示不与任何特定计算机、虚拟系统或者其它设备固有相关。各种 通用系统也可以与基于在此的示教一起使用。根据上面的描述，构造这类系统所要求的结构是显而易见的。此外，本公开也不针对任何特定编程语言。应当明白，可以利用各种编程语言实现在此描述的本公开的内容，并且上面对特定语言所做的描述是为了披露本公开的最佳实施方式。
在此处所提供的说明书中，说明了大量具体细节。然而，能够理解，本公开的实施例可以在没有这些具体细节的情况下实践。在一些实例中，并未详细示出公知的方法、结构和技术，以便不模糊对本说明书的理解。
类似地，应当理解，为了精简本公开并帮助理解各个公开方面中的一个或多个，在上面对本公开的示例性实施例的描述中，本公开的各个特征有时被一起分组到单个实施例、图、或者对其的描述中。然而，并不应将该公开的方法解释成反映如下意图：即所要求保护的本公开要求比在每个权利要求中所明确记载的特征更多的特征。更确切地说，如下面的权利要求书所反映的那样，公开方面在于少于前面公开的单个实施例的所有特征。因此，遵循具体实施方式的权利要求书由此明确地并入该具体实施方式，其中每个权利要求本身都作为本公开的单独实施例。
本领域那些技术人员可以理解，可以对实施例中的设备中的模块进行自适应性地改变并且把它们设置在与该实施例不同的一个或多个设备中。可以把实施例中的模块或单元或组件组合成一个模块或单元或组件，以及此外可以把它们分成多个子模块或子单元或子组件。除了这样的特征或过程或者单元中的至少一些是相互排斥之外，可以采用任何组合对本说明书(包括伴随的权利要求、摘要和附图)中公开的所有特征以及如此公开的任何方法或者设备的所有过程或单元进行组合。除非另外明确陈述，本说明书(包括伴随的权利要求、摘要和附图)中公开的每个特征可以由提供相同、等同或相似目的的替代特征来代替。
此外，本领域的技术人员能够理解，尽管在此所述的一些实施例包括其它实施例中所包括的某些特征而不是其它特征，但是不同实施例的特征的组合意味着处于本公开的范围之内并且形成不同的实施例。例如，在下面的权利要求书中，所要求保护的实施例的任意之一都可以以任意的组合方式来使用。
应该注意的是上述实施例对本公开进行说明而不是对本公开进行限制，并且本领域技术人员在不脱离所附权利要求的范围的情况下可设计出替换实施例。在权利要求中，不应将位于括号之间的任何参考符号构造成对权利要求的限制。单词“包含”不排除存在未列在权利要求中的元件或步骤。位于元件之前的单词“一”或“一个”不排除存在多个这样的元件。本公开可以借助于包括有若干不同元件的硬件以及借助于适当编程的计算机来实现。在列举了若干装置的单元权利要求中，这些装置中的若干个可以是通过同一个硬件项来具体体现。单词第一、第二、以及第三等的使用不表示任何顺序。可将这些单词解释为名称。

Claims (22)

  1. 一种电子书消费需求的确定方法，其包括：
    获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定所述阅读用户针对该电子书执行的各次交互操作的操作时间戳；
    根据所述各次交互操作中相邻的两次交互操作之间的时间间隔，确定所述阅读用户针对该电子书的交互时长；
    根据所述阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与所述阅读用户相对应的深度交互电子书；
    根据筛选出的深度交互电子书以及对应所述深度交互电子书的消费金额，确定所述阅读用户的消费需求信息。
  2. 根据权利要求1所述的方法，其中，所述根据所述各次交互操作中相邻的两次交互操作之间的时间间隔，确定所述阅读用户针对该电子书的交互时长的步骤具体包括：
    当确定所述各次交互操作中包含预设持续类操作时，结合所述预设持续类操作所对应的操作持续时长确定所述阅读用户针对该电子书的交互时长。
  3. 根据权利要求1或2所述的方法，其中，所述根据所述各次交互操作中相邻的两次交互操作之间的时间间隔，确定所述阅读用户针对该电子书的交互时长的步骤具体包括：
    分别针对每相邻的两次交互操作，根据该相邻的两次交互操作之间的时间间隔和预设的时间轴分片累加规则中的至少一个，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
    根据已记录的各个交互间隔确定所述阅读用户针对该电子书的交互时长。
  4. 根据权利要求3所述的方法，其中，根据该相邻的两次交互操作之间的时间间隔，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔的步骤具体包括：
    判断该相邻的两次交互操作之间的时间间隔是否大于预设间隔阈值；若否，将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。
  5. 根据权利要求3所述的方法，其中，根据预设的时间轴分片累加规则，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔的步骤具体包括：
    预先将所述各次交互操作按照操作时间戳记录到与该电子书相对应的交互时间轴上；
    按照所述预设的时间轴分片累加规则，将所述交互时间轴划分为多个时间分片；
    分别针对每个时间分片，将属于该时间分片的各次交互操作中每相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
    则所述根据已记录的各个交互间隔确定所述阅读用户针对该电子书的交互时长的步骤具体包括：
    分别针对每个时间分片，根据该时间分片内记录的各个交互间隔确定该时间分片的分片交互子时长，根据各个时间分片的分片交互子时长确定所述阅读用户针对该电子书的交互时长。
  6. 根据权利要求5所述的方法，其中，按照所述预设的时间轴分片累加规则，将所述时间轴划分为多个时间分片的步骤具体包括：
    根据预设分片时长和各次交互操作的操作时间戳中的至少一个，将所述时间轴划分为多个时间分片。
  7. 根据权利要求5或6所述的方法，其中，所述分别针对每个时间分片，根据该时间分片内记录的各个交互间隔确定该时间分片的分片交互子时长的步骤具体包括：
    分别针对每个时间分片，当确定属于该时间分片的各次交互操作中包含预设持续类操作时，进一步判断所述预设持续类操作的操作起始时间点与操作结束时间点是否同属于该时间分片；
    若否，根据所述操作起始时间点以及所述操作结束时间点计算操作持续时长；
    根据所述操作持续时长修正与该持续类操作相对应的至少两个时间分片的分片交互子时长。
  8. 根据权利要求1-7任一所述的方法，其中，所述获取与阅读用户发生交互的电子书的步骤具体包括：
    确定所述阅读用户的历史阅读记录中包含的各个电子书；
    按照预设剔除规则，对所述阅读用户的历史阅读记录中包含的各个电子书进行剔除处理；
    将执行所述剔除处理之后剩余的电子书确定为所述与阅读用户发生交互的电子书；
    其中，所述预设剔除规则包括以下中的至少一个：剔除用户未下载过的电子书、剔除预设时段内未发生交互的电子书、剔除免费试读章节的已读比例小于预设剔除比例的电子书、剔除免费的电子书、以及剔除用户已完整付费的电子书。
  9. 根据权利要求1-8任一所述的方法，其中，所述从获取到的电子书中筛选与所述阅读用户相对应的深度交互电子书的步骤具体包括：
    按照预设筛选规则，从获取到的电子书中筛选与所述阅读用户相对应的深度交互电子书；
    其中，所述预设筛选规则包括以下中的至少一个：
    筛选交互时长大于第一预设时长，且执行过搜索操作的电子书；
    筛选交互时长大于第二预设时长并且具有以下特征中的至少一种特征的电子书：已下载章节占整书章节的比例大于预设筛选比例、已下载章节的数量大于预设筛选数量、和已下载次数大于预设次数；
    筛选交互时长大于第三预设时长，且存在付费行为的电子书；以及，
    筛选交互时长大于第四预设时长的电子书。
  10. 根据权利要求1-9任一所述的方法，其中，所述根据筛选出的深度交互电子书以及对应所述深度交互电子书的消费金额，确定所述阅读用户的消费需求信息的步骤具体包括：
    进一步结合下述至少一项信息，对所述消费需求信息进行修正：
    所述阅读用户的历史日消费额、全网用户的平均日消费额、需求转化天数、以及预设的单日最大充值金额。
  11. 一种电子设备，包括：处理器、存储器、通信接口和通信总线，所述处理器、所述存储器和所述通信接口通过所述通信总线完成相互间的通信；
    所述存储器用于存放至少一可执行指令，所述可执行指令使所述处理器执行以下操作：
    获取与阅读用户发生交互的电子书，针对获取到的每本电子书，确定所述阅读用户针对该电子书执行的各次交互操作的操作时间戳；
    根据所述各次交互操作中相邻的两次交互操作之间的时间间隔，确定所述阅读用户针对该电子书的交互时长；
    根据所述阅读用户针对获取到的每本电子书的交互时长，从获取到的电子书中筛选与所述阅读用户相对应的深度交互电子书；
    根据筛选出的深度交互电子书以及对应所述深度交互电子书的消费金额，确定所述阅读用户的消费需求信息。
  12. 根据权利要求11所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    当确定所述各次交互操作中包含预设持续类操作时，结合所述预设持续类操作所对应的操作持续时长确定所述阅读用户针对该电子书的交互时长。
  13. 根据权利要求11或12所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    分别针对每相邻的两次交互操作，根据该相邻的两次交互操作之间的时间间隔和预设 的时间轴分片累加规则中的至少一个，确定是否将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
    根据已记录的各个交互间隔确定所述阅读用户针对该电子书的交互时长。
  14. 根据权利要求13所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    判断该相邻的两次交互操作之间的时间间隔是否大于预设间隔阈值；若否，将该相邻的两次交互操作之间的时间间隔记录为一个交互间隔。
  15. 根据权利要求13所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    预先将所述各次交互操作按照操作时间戳记录到与该电子书相对应的交互时间轴上；
    按照所述预设的时间轴分片累加规则，将所述交互时间轴划分为多个时间分片；
    分别针对每个时间分片，将属于该时间分片的各次交互操作中每相邻的两次交互操作之间的时间间隔记录为一个交互间隔；
    则所述根据已记录的各个交互间隔确定所述阅读用户针对该电子书的交互时长的步骤具体包括：
    分别针对每个时间分片，根据该时间分片内记录的各个交互间隔确定该时间分片的分片交互子时长，根据各个时间分片的分片交互子时长确定所述阅读用户针对该电子书的交互时长。
  16. 根据权利要求15所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    根据预设分片时长和各次交互操作的操作时间戳中的至少一个，将所述时间轴划分为多个时间分片。
  17. 根据权利要求15或16所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    分别针对每个时间分片，当确定属于该时间分片的各次交互操作中包含预设持续类操作时，进一步判断所述预设持续类操作的操作起始时间点与操作结束时间点是否同属于该时间分片；
    若否，根据所述操作起始时间点以及所述操作结束时间点计算操作持续时长；
    根据所述操作持续时长修正与该持续类操作相对应的至少两个时间分片的分片交互子时长。
  18. 根据权利要求11-17任一所述的电子设备，其中，所述可执行指令使所述处理器 执行以下操作：
    确定所述阅读用户的历史阅读记录中包含的各个电子书；
    按照预设剔除规则，对所述阅读用户的历史阅读记录中包含的各个电子书进行剔除处理；
    将执行所述剔除处理之后剩余的电子书确定为所述与阅读用户发生交互的电子书；
    其中，所述预设剔除规则包括以下中的至少一个：剔除用户未下载过的电子书、剔除预设时段内未发生交互的电子书、剔除免费试读章节的已读比例小于预设剔除比例的电子书、剔除免费的电子书、以及剔除用户已完整付费的电子书。
  19. 根据权利要求11-18任一所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    按照预设筛选规则，从获取到的电子书中筛选与所述阅读用户相对应的深度交互电子书；
    其中，所述预设筛选规则包括以下中的至少一个：
    筛选交互时长大于第一预设时长，且执行过搜索操作的电子书；
    筛选交互时长大于第二预设时长并且具有以下特征中的至少一种特征的电子书：已下载章节占整书章节的比例大于预设筛选比例、已下载章节的数量大于预设筛选数量、和已下载次数大于预设次数；
    筛选交互时长大于第三预设时长，且存在付费行为的电子书；以及，
    筛选交互时长大于第四预设时长的电子书。
  20. 根据权利要求11-19任一所述的电子设备，其中，所述可执行指令使所述处理器执行以下操作：
    进一步结合下述至少一项信息，对所述消费需求信息进行修正：
    所述阅读用户的历史日消费额、全网用户的平均日消费额、需求转化天数、以及预设的单日最大充值金额。
  21. 一种非易失性计算机可读存储介质，所述非易失性计算机可读存储介质中存储有至少一可执行指令，所述可执行指令用于使该计算机执行前述任一权利要求1-10所述的电子书消费需求的确定方法。
  22. 一种计算机程序产品，该计算机程序产品包括存储在非易失性计算机可读存储介质上的计算程序，该计算机程序包括程序指令，当该程序指令被计算机执行时，使该计算机执行前述任一权利要求1-10所述的电子书消费需求的确定方法。
PCT/CN2019/073492 2018-03-30 2019-01-28 电子书消费需求的确定方法、电子设备及计算机存储介质 WO2019184582A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810299147.3 2018-03-30
CN201810299147.3A CN108520440B (zh) 2018-03-30 2018-03-30 阅读用户的总消费存量需求的确定方法、电子设备及计算机存储介质

Publications (1)

Publication Number Publication Date
WO2019184582A1 true WO2019184582A1 (zh) 2019-10-03

Family

ID=63431941

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/073492 WO2019184582A1 (zh) 2018-03-30 2019-01-28 电子书消费需求的确定方法、电子设备及计算机存储介质

Country Status (2)

Country Link
CN (1) CN108520440B (zh)
WO (1) WO2019184582A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108520440B (zh) * 2018-03-30 2020-12-08 掌阅科技股份有限公司 阅读用户的总消费存量需求的确定方法、电子设备及计算机存储介质
CN109299350B (zh) * 2018-09-13 2019-08-20 掌阅科技股份有限公司 电子书的排序方法、计算设备及计算机存储介质
CN109241149B (zh) * 2018-09-13 2019-08-30 掌阅科技股份有限公司 电子书的排序方法、计算设备及计算机存储介质
CN109919719A (zh) * 2019-02-22 2019-06-21 掌阅科技股份有限公司 信息推送方法、计算设备及计算机存储介质
CN112927053B (zh) * 2021-03-31 2022-02-22 掌阅科技股份有限公司 书籍阅读权限的处理方法、计算设备及计算机存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104081428A (zh) * 2011-12-27 2014-10-01 英特尔公司 用于生成和分析数字可读媒体消费数据的方法、设备和系统
CN105308587A (zh) * 2013-06-14 2016-02-03 微软技术许可有限责任公司 将可消费内容的用户使用结合于推荐中
CN105677845A (zh) * 2016-01-06 2016-06-15 北京京东尚科信息技术有限公司 电子书籍的推送方法和装置
CN106777139A (zh) * 2016-12-19 2017-05-31 浙江工业大学 基于阅读时间的用户阅读偏好统计的个性化推送方法
CN108520440A (zh) * 2018-03-30 2018-09-11 掌阅科技股份有限公司 电子书消费需求的确定方法、电子设备及计算机存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001265481A (ja) * 2000-03-21 2001-09-28 Nec Corp ページ情報表示方法及び装置並びにページ情報表示用プログラムを記憶した記憶媒体
US8135696B2 (en) * 2007-03-03 2012-03-13 Michael John Safoutin Time-conditioned search engine interface with visual feedback
CN104599152A (zh) * 2014-03-15 2015-05-06 北京智膜科技有限公司 基于智能信息设备多层次输入界面的信息采集及处理方法
CN104090741A (zh) * 2014-06-17 2014-10-08 小米科技有限责任公司 电子书阅读统计方法及装置
CN107491252B (zh) * 2017-08-21 2018-05-29 掌阅科技股份有限公司 阅读时长的统计方法及电子设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104081428A (zh) * 2011-12-27 2014-10-01 英特尔公司 用于生成和分析数字可读媒体消费数据的方法、设备和系统
CN105308587A (zh) * 2013-06-14 2016-02-03 微软技术许可有限责任公司 将可消费内容的用户使用结合于推荐中
CN105677845A (zh) * 2016-01-06 2016-06-15 北京京东尚科信息技术有限公司 电子书籍的推送方法和装置
CN106777139A (zh) * 2016-12-19 2017-05-31 浙江工业大学 基于阅读时间的用户阅读偏好统计的个性化推送方法
CN108520440A (zh) * 2018-03-30 2018-09-11 掌阅科技股份有限公司 电子书消费需求的确定方法、电子设备及计算机存储介质

Also Published As

Publication number Publication date
CN108520440B (zh) 2020-12-08
CN108520440A (zh) 2018-09-11

Similar Documents

Publication Publication Date Title
WO2019184582A1 (zh) 电子书消费需求的确定方法、电子设备及计算机存储介质
CN106294866B (zh) 一种日志处理方法和装置
CN109657138B (zh) 一种视频推荐方法、装置、电子设备及存储介质
CN106407190B (zh) 一种事件记录查询方法及装置
WO2019184583A1 (zh) 基于电子书的活动内容推送方法及电子设备
RU2636133C2 (ru) Способ и устройство для отображения прикладного программного обеспечения
CN110046929B (zh) 一种欺诈团伙识别方法、装置、可读存储介质及终端设备
CN107689008A (zh) 一种用户投保行为预测的方法及装置
US20170364506A1 (en) System and method for generating phrase based categories of interactions
CN110489646B (zh) 用户画像构建方法及终端设备
CN109543940B (zh) 活动评估方法、装置、电子设备及存储介质
CN110020773A (zh) 一种数据处理方法、设备及计算机可读介质
CN110866698A (zh) 用于评定服务提供方的服务分值的装置
CN110599240A (zh) 应用的偏好值确定方法、装置和设备及存储介质
CN107749888A (zh) 一种用户行为数据的统计方法
CN111179090A (zh) 一种交易数据处理方法、装置及设备
CN108428138B (zh) 一种基于客户分群的客户生存率分析装置及分析方法
US10600063B2 (en) Real-time system to identify and analyze behavioral patterns to predict churn risk and increase retention
Pederzoli Stochastic volatility and GARCH: A comparison based on UK stock data
US20150199781A1 (en) Patent Valuation System
CN110717653A (zh) 风险识别方法及装置和电子设备
US10560057B1 (en) Measuring media attention over time based on long term heterogeneous archive data
CN110737691B (zh) 用于处理访问行为数据的方法和装置
CN106156122B (zh) 交易信息获取方法及装置
CN109063015B (zh) 热点内容的提取方法、装置及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19777335

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19777335

Country of ref document: EP

Kind code of ref document: A1