CN106621328B - A kind of game role behavioral data processing method and system - Google Patents
A kind of game role behavioral data processing method and system Download PDFInfo
- Publication number
- CN106621328B CN106621328B CN201510741360.1A CN201510741360A CN106621328B CN 106621328 B CN106621328 B CN 106621328B CN 201510741360 A CN201510741360 A CN 201510741360A CN 106621328 B CN106621328 B CN 106621328B
- Authority
- CN
- China
- Prior art keywords
- player
- view
- node
- same name
- game
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses a kind of game role behavioral data processing method and system, the processing methods are as follows: converts bottom data for the data information in game data log first；Then player's footprint view, node of the same name analysis view are generated and is lost according to the bottom data and play furniture body Track View；Player's footprint view is used to show the whole trend of all player's footprints, and the conversion ratio index between each step；The node analysis view of the same name is used to show the arrival number of the node of the same name, exits number and exits rate；Furniture body Track View is played in the loss, leaves the player of game specific track experienced before node of the same name herein for showing from any node of the same name.Visualization technique is applied in the processing of game role behavioral data, so as to effectively determine the behavior node for making game experiencing generate obstacle, and a kind of fabulous approach for improving game design can be provided to avoid the dependence to priori knowledge.
Description
Technical field
The present invention relates to game data processing technology field more particularly to a kind of game role behavioral data processing method and
System.
Background technique
Successful game products, the requirement of technology and user experience to product itself are all very high.And product
Innovation and improve and all greatly rely on data, many qualitative evaluation methods, as VIP test, questionnaire, face or
The methods of recorded video is employed to guarantee the interest of game products and the balance of sports.In addition to direct from game forum
It obtains except player feedback, another kind of feedback is exactly game data log, it gives we provide research player's behavior, is changed
The fabulous approach of kind game design.
For at present, Statistics-Based Method is widely used in game data analysis.However, they are more or less
It is required that there are some priori knowledges to data, but the mode in game data, can not predict in advance.In addition, for product sheet
The case where body, macroscopical index explanation is only capable of embodying game substantially, any part can not be still determined sometimes based on these data
Improved.Visualization is the method for a kind of effective analysis and display data, its main purpose is by graphical
Clear, effective information transmitting that means carry out.
How visualization technique to be applied in the processing of game role behavioral data to and be avoided the dependence to priori knowledge,
Technical support is provided for the innovation and improvement of development of games, is badly in need of researching and solving in various game technologies.
The disclosure of background above technology contents is only used for auxiliary and understands inventive concept and technical solution of the invention, not
The prior art for necessarily belonging to present patent application shows above content in the applying date of present patent application in no tangible proof
In the case where having disclosed, above-mentioned background technique should not be taken to the novelty and creativeness of evaluation the application.
Summary of the invention
It is an object of the invention to propose a kind of game role behavioral data processing method, deposited with solving the above-mentioned prior art
Be unable to effective position improve position and rely on priori knowledge the technical issues of.
For this purpose, the present invention proposes a kind of game role behavioral data processing method, first by the number in game data log
It is believed that breath is converted into bottom data；Then according to the bottom data generate player's footprint view, node of the same name analysis view and
It is lost and plays furniture body Track View；Player's footprint view is used to show the whole trend of all player's footprints, and respectively walks it
Between conversion ratio index；The node analysis view of the same name is used to show the arrival number of the node of the same name, exits number and exit
Rate；Furniture body Track View is played in the loss, for showing that the player for leaving game from any node of the same name is of the same name herein
Specific track experienced before node.
According to embodiment, processing method of the invention may also include following technical characteristic:
Player's footprint view supports user's interaction, checks player's footprint situation of change to facilitate；The loss player
Specific Track View supports user's interaction, for checking the concrete behavior track and/or each concrete behavior track of each node player
Accounting.
Player's footprint view, which is additionally provided with, shows the adjustable of player's footprint varied number for controlling with selectivity
First threshold.
In the node of the same name analysis view to the arrivals number, exit several and exit rate and be ranked up respectively.
Deduplication operation also is carried out to player in the node analysis view of the same name.
Temporally inverted order is shown for the concrete behavior track for being lost each node player in object for appreciation furniture body Track View.
The present invention also proposes a kind of game role behavioral data processing system, including data information conversion module, and being used for will
Data information in game data log is converted into bottom data；View generation module, for being distinguished according to the bottom data
It generates player's footprint view, node of the same name analysis view and is lost and play furniture body Track View, wherein the view generation module
It include: that furniture body track view is played in player's footprint view generation submodule, node of the same name analysis view generation submodule and loss
Figure generates submodule；Player's footprint view generation submodule shows all player's footprints for generating player's footprint view
Whole trend, and the conversion ratio index between each step；The node analysis view generation submodule of the same name is of the same name for generating
Node analyzes view, shows the arrival number of the node of the same name, exits number and exits rate；Furniture body Track View is played in the loss
Submodule is generated, furniture body Track View is played for generating to be lost, shows the player for leaving game from any node of the same name
Specific track experienced before node of the same name herein.
According to embodiment, processing system of the invention may also include following technical characteristic:
Player's footprint view generation submodule further includes interactive unit, for supporting user's interaction, is checked with facilitating
Player's footprint situation of change；It further includes interactive unit that the loss, which plays furniture body Track View and generates submodule, is used for supporting
Family interaction, and check the concrete behavior track of each node player and/or the accounting of each concrete behavior track.
Player's footprint view generation submodule further includes that first threshold adjusts unit, for controlling with selective display
Player's footprint varied number.
The node of the same name analysis view generation submodule further includes sequencing unit, for the arrival number, exit number
And it exits rate and is ranked up respectively.
Duplicate removal unit is still further comprised in the node analysis view generation submodule of the same name, for going to player
It operates again.
Temporally inverted order is shown for the concrete behavior track for being lost each node player in object for appreciation furniture body Track View.
The loss, which is played furniture body Track View and generated, to be additionally provided with second threshold and adjusts unit in submodule, for adjust with
Selectivity shows the concrete behavior track of different accountings.
The beneficial effect of the present invention compared with the prior art includes: that visualization technique is applied to game role behavioral data
In processing, so as to effectively determine the behavior node for making game experiencing generate obstacle, and can be to avoid to priori knowledge
It relies on, provides a kind of fabulous approach for improving game design.
Detailed description of the invention
Fig. 1 is the flow diagram of the processing method of the embodiment of the present invention one；
Fig. 2 is the structural block diagram of the processing system of the embodiment of the present invention two；
Fig. 3 is player's footprint view screenshot in a specific embodiment；
Fig. 4 is the screenshot of the node of the same name analysis view and loss object for appreciation furniture body Track View in a specific embodiment
(wherein, top is that node of the same name analyzes view, and lower part is to be lost to play furniture body Track View)；
Fig. 5 a, 5b, 5c are the behavior sequence schematic diagram of three players in a specific embodiment respectively；
Fig. 6 a, 6b, 6c be in a specific embodiment respectively according to reach number, exit number and exit rate sequence it is visual
Change the screenshot of view；
Fig. 7 a, 7b, 7c be in the specific embodiment of corresponding diagram 6a, 6b, 6c to after player's number duplicate removal respectively according to arriving
Up to number, the screenshot for exiting number and exiting the visualization view that rate sorts；
Fig. 8 a is player's footprint view screenshot an of specific embodiment, view when Fig. 8 b is a node in suspension Fig. 8 a
Figure screenshot；
Fig. 9 a is the screenshot for being adjusted to loss when only showing two kinds of track behaviors and playing furniture body Track View, and Fig. 9 b is to adjust
The screenshot of furniture body Track View is played in loss when section is all shown to all footprint behaviors.
Specific embodiment
The invention proposes a kind of game role behavioral data processing method and system based on visualization analysis technique, can
Player may when user being helped to understand the Behavior law, especially game new hand stage of player's behavior, monitoring player in gaming
Loss retain point, provide reference for the further iteration of product.Meanwhile from player grade the specific loss of data layer auto-tracking player
Specific action trail has found that it is likely that the behavior node for go deep into player game experiencing generation obstacle, to help game to set
Count the fluency of Shi Gaijin game experiencing.
In gaming, being analyzed using funnel model has had certain methods and case, tries to understand player to play
The real purpose of game provides reasonable access path and new hand's boot flow etc. for them, rather than goes to improve simply
Conversion ratio.Funnel model is usually used in the turnover rate of each step of analyzing web site process, and is shown with funnel model.It analyzed
Journey can be substantially unfolded from following three step: 1) the determining access path or operating process for needing to analyze, and 2) collect data simultaneously
Count the number of each step in the path respectively, 3) result is shown with funnel model.
In web analytics, the click steam of user each time is exactly the access path of user.In most cases
The access path of user be arbitrarily it is unordered, user when access one website may commonly using retreat, return homepage or
Some link etc. is clicked directly on, the registration of different user access path may only have 1%, and analyzing these unordered paths is not have
Significant.Therefore, funnel model is used in web analytics, and primarily directed to some critical paths, i.e. user is for some
Target and the orderly path for entering a relative standard, the target of user be exactly in order to reach " outlet ", rather than it is random
It goes around.Register flow path, shopping process such as e-commerce website, service, process for using of applied website etc..Thus it is possible to
The conversion ratio of each step is calculated according to these critical paths.
In funnel model, conversion ratio is exactly the number ratio for entering next page from a page, such as access head
The user of page has 30, and has 12 from the user that homepage clicks to enter lower one page, then the conversion ratio from homepage to next page is just
It is 12/30=40%.Of course, it is possible to according to the access path of user calculate each page to the next page conversion ratio, but this
Class work is not largely nonsensical for targetedly website for, but needs to concentrate on crucial points, that is, extracts pass
The conversion ratio in key path is just significant.By the statistics to these data and index, the process conversion ratio can be tentatively judged
The turnover rate situation of situation and each step.Next, funnel model need to be used in order to allow the more specific image of result of analysis
Intuitively shown.
The shortcomings that based on funnel model analysis:
Funnel model not only shows user and is entering final conversion ratio of the process to realization target, while can also show
The conversion ratio of each step in critical path.But single funnel model is little for analysis significance, because being unable to Dan Congyi funnel
In model evaluate some critical workflow of website in each step conversion ratio quality, so must by trend, compare and segment
Method the conversion ratio of step each in process is analyzed.
Trend: being analyzed from the situation of change of time shaft, suitable for changing to a certain process or in which some step
Into or optimization effect monitoring.
Compare: by comparing the conversion ratio of purchase or process for using between similar products or service, finding certain products or answer
The problem of using.
Subdivision: performance of the subdivision source or different customer types on conversion ratio, find some high quality source or
Client, commonly used in the advertisement of analyzing web site or promotion effect etc..
Therefore, funnel model is relatively specific for the analysis of the conversion ratio of certain critical paths in website, therefore, for game
For behavioural analysis, need to define these critical paths in advance, and for different game, critical path is different
, it is so more huge to artificial bring treating capacity.When analyzing a game, especially development of games, test
Initial stage is especially envisaged that the loss of game and retains phenomenon.In index Design, there is next day to retain, retain, stay within 7th within 3rd
Rate etc. is deposited, also has the comparison between popularization measurement and game transverse direction of the similar index for game in loss, these refer to
Target really plays their irreplaceable roles.The reason of game behavior of player is multifarious, and player is lost is also due to people
It is different, but how player is lost on earth? where it is lost? the present invention is based on visualization analysis techniques, propose a kind of logical
With (" general " refers to that the analytical mathematics in terms of game data analysis are general, and traditional funnel model needs specific definition a certain
Access path, and analyzed using funnel model, and the present invention completely covers a variety of different access path, therefore phase
Than traditional funnel model, the present invention is more comprehensively complete.) game role behavioral data processing method and system, can be complete
The various aspects of funnel model are covered, meanwhile, the analysis in each path, the analysis of each node are more fully goed deep into, from entirety
The specific track details of overview to player have detailed analysis, can satisfy the game role of this special technical field of game
Behavioural analysis demand.
The present invention solves the problems, such as following aspects:
1) play stage which be may cause player loss key node: for a game, position its
Life stage, the especially point of loss in new hand's stage, have very important significance for keeping player here.
2) player's path analysis experienced before loss: player before leaving a game, not necessarily last
It is a to operate the loss for leading to the player, it is also possible to caused by the operation or bad game experiencing before the player, therefore, to have
Necessity analyzes specific action trail in the level of player's grade data.
3) in terms of versatility, it is necessary to carry out lateral comparison to similar game products, understand in similar object for appreciation domestic discipline and family rules
On mould, the situation of change of the conversion ratio of each step: how player gos deep into degree to a new game of trial, turns in which step
There is biggish loss when change.
The above technical problem both includes service layer, also there is the insurmountable technology barrier (funnel model such as funnel model
Etc. the access path for needing predefined to need to analyze, and in specific game, the behavior of player is multifarious, in addition to some spies
Outside other predominating path, other paths are difficult to go to be analyzed by predefined).The present invention provides a kind of general to lead to
It crosses games log data quickly to analyze role-act transition, be lost the method and system for retaining positioning, trajectory analysis etc., completely cover
A variety of different access path have been covered, the insurmountable technical problem such as conventional funnel model is overcome.
Embodiment one:
As shown in Figure 1, being the stream of game role behavioral data processing method of the present embodiment based on visualization analysis technique
Cheng Tu, this method are as follows: convert bottom data for the data information in game data log first；Then, according to the bottom
Data, which generate player's footprint view, node of the same name analysis view and are lost, plays furniture body Track View.
Wherein, player's footprint view is used to show the whole trend of all player's footprints, and the conversion between each step
Rate index；The node analysis view of the same name is used to show the arrival of the node of the same name (node for referring to identical game behavior)
Number exits number and exits rate；Furniture body Track View is played in the loss, leaves game for showing from any node of the same name
Player specific track experienced before node of the same name herein.
Player's footprint situation of change is checked for convenience, and player's footprint view is set as with the function for supporting user's interaction
Energy.Player's footprint view is additionally provided with for controlling adjustable first threshold for showing player's footprint varied number with selectivity
Value, can simplify view by the adjusting of the first threshold and shows by different level, and exclude to occur individually does not have practical meaning
The footprint situation of change etc. of justice.
The concrete behavior track of each node player of the same name and/or the accounting of each concrete behavior track are checked for convenience, it is described
It is lost object for appreciation furniture body Track View to be set as supporting user's interaction, that is, there is integration of user interaction functionality.Furniture body track is played in the loss
Temporally inverted order is shown for the concrete behavior track of each node player in view, i.e., according to 1 walking last of last reactive power optimization-
It is for -1 walking second from the bottom ..., and so on.The loss, which is played, can also configure in furniture body Track View for selectivity
Show adjustable second threshold of the concrete behavior track of different accountings.
In the node of the same name analysis view to the arrivals number, exit several and exit rate and be ranked up respectively.It is described same
Deduplication operation also is carried out to player in reputation and integrity point analysis view, (brush copy is game because certain player brushes some game transcript repeatedly
The generic term in field: entering game transcript, and the tasks backed off after random such as strange is beaten in completion) when will lead to a large amount of node repetition, meeting
It affects to sequence.
Embodiment two:
The present invention also corresponds to previous embodiment one and provides a kind of game role behavioral data processing system, as shown in Fig. 2, packet
Data information conversion module is included, for converting bottom data and view generation for the data information in game data log
Module plays furniture body for generating player's footprint view, node of the same name analysis view respectively according to the bottom data and being lost
Track View.Wherein, the view generation module includes: player's footprint view generation submodule, node of the same name analysis view life
At submodule and it is lost object for appreciation furniture body Track View generation submodule.
Wherein, player's footprint view generation submodule shows all player's footprints for generating player's footprint view
Whole trend, and the conversion ratio index between each step.The node analysis view generation submodule of the same name is of the same name for generating
Node analyzes view, shows the arrival number of the node (node for referring to identical game behavior) of the same name, exits number and exits rate.Institute
It states loss object for appreciation furniture body Track View and generates submodule, be lost object for appreciation furniture body Track View for generating, show from any described
Node of the same name leaves the player of game specific track experienced before node of the same name herein.
Player's footprint view generation submodule further includes interactive unit, for supporting user's interaction, is checked with facilitating
Player's footprint situation of change.Player's footprint view generation submodule can also further comprise that first threshold adjusts unit, use
Player's footprint varied number is shown with selectivity in control.Adjusting unit by the first threshold can simplify view and by different level
It has been shown that, and exclude the footprint situation of change etc. without practical significance occurred individually.
It further includes interactive unit that the loss, which plays furniture body Track View and generates submodule, for supporting user's interaction, and
Check the concrete behavior track of each node player and/or the accounting of each concrete behavior track.Furniture body track view is played in the loss
Preferably temporally inverted order is shown for the concrete behavior track of each node player in figure, i.e., according to last reactive power optimization -1 step last
Behavior -1 walking second from the bottom is ..., and so on.The loss is played to be additionally provided in furniture body Track View generation submodule
Second threshold adjusts unit, for adjusting the concrete behavior track for showing different accountings with selectivity.
The node of the same name analysis view generation submodule further includes sequencing unit, for the arrival number, exit number
And it exits rate and is ranked up respectively.Duplicate removal unit is still further comprised in the node analysis view generation submodule of the same name, is used
In to player's progress deduplication operation.
Embodiment three:
The present embodiment is by a specific game and compares several attached drawings to do further detailed description as follows:
Firstly, there are three main views, player's footprint view, node of the same name analysis view and loss players in the present embodiment
Specific Track View, it is as shown in Figure 3 and Figure 4 respectively.After data load, player's footprint view can show institute in data first
There is the whole trend of player's behavior footprint, and the indexs such as conversion ratio between each step, and supports user's interaction abundant, so that
User can arbitrarily check player's behaviorism theory between two step of front and back (behaviorism theory namely footprint described previously variation) situation.Together
Reputation and integrity point analysis view by the arrival number for showing each game node of player's experience, exit number and exit rate etc., help, which positions, moves back
The higher node of the same name of extracting rate, quickly to position potential problems place.It is lost and plays furniture body Track View, for from a certain node
It leaves the player of the game game track experienced behavior before the node to analyze, assisting in may in process design
There are the problem of.
Behavior and the funnel model of game player the difference is that: the path that funnel model is portrayed is more single
, and this single path is also the interests conversion pathway that businessman pays close attention to the most.And game player's behavior path, initially into
Can be more similar when entering a game, and will break up after completing new hand's task very severe, the behavior of specific player
It will be multifarious.The data of the present embodiment analysis are games log data when triggering certain behaviors based on record player, such as
Table name is called the table of " src_login_day ", and behavior when what it was recorded is players log game, we extract players log
Behavior, player's behavior is named as " login ".For another example a table name is called the table of " src_pvp_day ", and what it was recorded is to play
Family carries out related data when PVP (player and player fight, Player VS Player) playing method, we then extract player
The behavioral data of PVP is named as " PVP ".In this way, to analysis in need player's user behaviors log carry out similar operations and
Behavior is extracted, in this way more than all log behavior tables progress for being related to player's behavior of daily whole players or newly-increased player
After operation, UNIONALL union operation is carried out, then successively sort by behavior triggering moment, obtain the behavior sequence of each player.
Next, the behavior sequence of all players is integrated, calculate the behavior that same day player is all in log be as
What transition, operation amount, behavior including each behavior of every single stepping and the varied number between behavior etc..Citing:
Assuming that there are three players, respectively as shown in Fig. 5 a, 5b, 5c, Wan Jiayi are as follows: log in-publishing-is logged in-is stepped on the behavior sequence of player 1
Out, player 2 are as follows: log in-playing method-is published-log in-playing method-is published-and log in-publish, and player 3 are as follows: log in-playing method-publishes.Structure
After having made these three behavior sequence lists, summarize calculating to each step, i.e. the first step logs in the players number of behavior and is
3, the players number that second step occurs to play judicial act is 2, and the players number for publishing behavior is 1, and so on, until having summarized most
That corresponding step of long sequence.Some behaviors may repeat, such as playing method-playing method-playing method etc., can be by prior
It defines one and ignores sequence, some behavior continuously occurred included in this sequence is merged into operation, becomes a row
For.Meanwhile also can record from a behavior and being transformed into the average duration that another behavior is spent, it can for analyzing and showing
The place of the time unreasonable distribution of energy.Complete be in by data structure using visualization technique after basic Data Integration summarizes
In present player's footprint view, citing is hereinafter illustrated to a drafting process of player's footprint view.Institute in the present embodiment
Data structure is stated to be exemplified below:
Dateid: date
Timestep: step number number
Action_id_before: preamble footprint type
Action_id_after: postorder footprint type
Action_change_num: variable quantity (how many player is transferred to postorder footprint from preamble footprint)
Action_number_before: the number of players of preamble footprint
Action_number_after: the number of players of postorder footprint
Action_average_time: average time-consuming (second) from preamble footprint to postorder footprint.
In player's footprint view, each step is the behavior set that player triggered, due to the tool of different players
Body footprint is differentiated, therefore the behavior set walked between step has duplicate node of the same name.In player's footprint figure, play
Family reaches the node simply by different steps, it is therefore necessary to Macro or mass analysis is carried out to these nodes of the same name, in this way, can
The specific footprint of player is observed to fine granularity, also can be from whole angle, the node that positioning is lost rapidly.
Node analysis view of the same name provides each behavior node (representing every kind of different game behavior) under whole step numbers and arrives
Up to number ,/exit the rate of count/exiting (exit rate=exits number/arrival number) analysis, and support three kinds of sortords: by reaching number row
Sequence, by exiting, number sorts and by rate sequence is exited, three kinds of ranking results in the present embodiment are respectively as shown in Fig. 6 a, 6b and 6c.
In practical business, some behavior nodes may be carried out repeatedly, as player may some game transcript of scrubbing brush repeatedly
Deng, therefore, player can be numbered and carry out duplicate removal, and be sorted accordingly, to obtain more meaningful ranking results, Fig. 7 a,
7b and 7c respectively corresponds Fig. 6 a, 6b and 6c and gives the node of the same name being ranked up before and after player's number duplicate removal by each sortord
Analyze view.
Player, which leaves game, must be not necessarily all that the last one behavior or operation cause player to have not there are many reason
Good game experiencing, so that player does not continue deeper into game, it is also possible to several steps before being player it is comprehensive
Conjunction game experiencing is bad and causes to be lost.Accordingly, it is determined that several foots of the player for exiting or being lost from a certain node before leaving
The distribution situation of mark is necessary.
It obtains first having to calculate the longest step number of all players, base after player's behavior transition (footprint variation) summarize data
It is successively drawn in this.A kind of each player's behavior of node on behalf, junction curve and its thickness representative between node have more
Few player is transformed into another endpoint (i.e. another behavior) from an endpoint (i.e. some behavior) for curve.Suspend some node
When, show exited from the node role's number (defined herein exits, refer to the players number for completing to publish after this node with
Direct the sum of the players number exited and the same day does not log in again), and show that exits simultaneously from the node exits rate and arrival rate.
The rate of exiting is referred to from the ratio of this node (directly publishing+role's number summation for not logging in again of the same day) and this angle of rotation of joint chromatic number
Value, arrival rate refer to the child's hair twisted in a knot-childhood chromatic number of this node and the ratio of first node (just having started whole players numbers).When clicking node
Show the conversion meaning that latter linked curve represents before the node.In addition, needing to control transition behavior to simplify overview
The first threshold of quantity is controlled by the first threshold and shows those mainstream behavior transition, without show scattered it is other
Transition behavior.In addition, the function of threshold value interactive adjustment, the level that user's control can be facilitated to show.Concretely: in interactive interface
One slider bar of upper offer, the threshold value as controlling behavior transition quantity adjusts operating knob, when minimum value (i.e. 1) is arrived in adjusting,
All behavior transition can all be shown；Otherwise, only show that those transition quantity are higher than the behavior transition of this threshold value.
In addition, in order to which player (player exited is lost player) the accounting feelings exited from each node of the same name are intuitively presented
The player's accounting situation exited is indicated that such user can identify rapidly and exit accounting with roundlet in each node center by condition
Biggish behavior node.In the lower section of footprint view, draws out a group arrival rate curve and (regard each step as a group, calculate
The accounting of the number of players that each step is retained and all number of players most started, the number of players that each step is retained with most open
The accounting of all number of players to begin is drawn with curve), two step of front and back carries out arrival rate contrast difference, can show in this way
Which step player more, also to facilitate user's positioning that may go wrong node is left out.
In addition, in order to analyze the game action trail from the player that each node leaves before this node, we also by
The data for storing player rank, track data of the player before each node leaves are come out with Drawing of Curve, as Fig. 8 a gives
The effect of visualization of player's footprint view.For example, when suspension " literary Wang Zhouyi ", screenshot as shown in Figure 8 b is obtained, can be seen
It is 4 to the role's number exited from the node, the rate that exits is 0.67%, and arrival rate is 61.94%；When clicking the node, track L1
(see Fig. 8 a) represent exited from the node and the same day do not have subsequent behavior player specific track, the thicknesses of lines represents from this
Player's accounting that paths exit.As shown in Figure 8 a, the specific of the player for not having subsequent behavior was exited from " literary Wang Zhouyi " same day
Track is to log in-start from scratch 1- to start from scratch 2- text Wang Zhouyi 1- text Wang Zhouyi 2, and the meaning of the track is to provide the user
Analysis is exited and does not have how the player of subsequent behavior exits, is which game behavior backed off after random specifically completed
, carrying out display to the specific track of these players can be used for assistant analysis.
It is played in furniture body Track View being lost, checks the specific action trail of player exited from each node, can chase after
The behavior that track player mainly undergoes before leaving game may cause the reason of it leaves convenient for customer analysis.Such as to check from
The specific track for the player that " keep a good distance from 1 " exits and each track accounting situation can click " moving back in node analysis view of the same name
Three houses 1 " are kept away, the concrete behavior track of the player exited from behavior node is subsequently generated.Mouse is suspended on each node, display
The title of the node, the specific action trail of player are that inverted order is shown, i.e., last reactive power optimization-inverse step 1 behavior-inverse
Step 2 behavior etc..In addition, for the purposes of the clarity of display, it is configured with the threshold value slider bar that control track show, is used for pair
Adjustable second threshold is adjusted to filter the lower action trail of accounting.As shown in Fig. 9 a, 9b, moved back from behavior node
For player out altogether there are two types of footprint behavior, most thick footprint behavior (as shown in L2 in Fig. 9 a) has 2 players to exit, when the
When two threshold values are transferred to minimum, all footprint behaviors (as shown in L3 in Fig. 9 b) all show out.
Before using technical solution provided by the invention, Game analysis personnel only reside within respectively to retention analysis is lost
Kind index is built, or searches critical path using funnel model, can not move towards analyze to the footprint of player's entirety,
Potential loss behavior node can not be quickly positioned, the present invention is provided for the intuitive approach of processing analysis, it is known that
Player is in the distribution situation of each behavior node, and convenient for positioning the position where possible defect, being conducive to subsequent game improves.
It would be recognized by those skilled in the art that it is possible for making numerous accommodations to above description, so embodiment is only
For describing one or more particular implementations.It will be understood by those skilled in the art that it can be variously modified and is replaced,
Without departing from spirit of the invention.Furthermore it is possible to many modifications are made so that specific condition to be fitted to design of the invention, and
Without departing from invention described herein central concept.So the present invention is not only restricted to specific embodiment disclosed herein, it may
It further include all embodiments and its equivalent for belonging to the scope of the invention.
Claims (14)
1. a kind of game role behavioral data processing method, it is characterised in that: first by the data information in game data log
It is converted into bottom data；Then player's footprint view, node of the same name analysis view are generated and is lost according to the bottom data and played
Furniture body Track View；
Player's footprint view is used to show the whole trend of all player's footprints, and the conversion ratio index between each step；
The node analysis view of the same name is used to show the arrival number of the node of the same name, exits number and exit rate, described of the same name
Node refers to the node that identical game behavior is carried out by player；
Furniture body Track View is played in the loss, for showing that the player for leaving game from any node of the same name is of the same name herein
Specific track experienced before node.
2. game role behavioral data processing method as described in claim 1, it is characterised in that: player's footprint view branch
User's interaction is held, checks player's footprint situation of change to facilitate；The loss plays furniture body Track View and supports user's interaction, uses
Accounting in the concrete behavior track and/or each concrete behavior track of checking each node player.
3. game role behavioral data processing method as claimed in claim 2, it is characterised in that: player's footprint view is also
Equipped with for controlling the adjustable first threshold for showing player's footprint varied number with selectivity.
4. game role behavioral data processing method as described in claim 1, it is characterised in that: the node analysis view of the same name
In figure to the arrival number, exit number and exit rate and be ranked up respectively.
5. game role behavioral data processing method as described in claim 1, it is characterised in that: the node analysis view of the same name
Deduplication operation also is carried out to player in figure.
6. game role behavioral data processing method as described in claim 1, it is characterised in that: furniture body rail is played in the loss
Temporally inverted order is shown for the concrete behavior track of each node player in mark view.
7. game role behavioral data processing method as claimed in claim 2, it is characterised in that: furniture body rail is played in the loss
Adjustable second threshold that the concrete behavior track of different accountings is shown for selectivity is also configured in mark view.
8. a kind of game role behavioral data processing system, characterized by comprising: data information conversion module is used for game
Data information in data logging is converted into bottom data；View generation module, for being generated respectively according to the bottom data
Furniture body Track View is played in player's footprint view, node of the same name analysis view and loss, wherein the view generation module packet
Include: furniture body Track View is played in player's footprint view generation submodule, node of the same name analysis view generation submodule and loss
Generate submodule；
Player's footprint view generation submodule shows that the entirety of all player's footprints is walked for generating player's footprint view
To, and the conversion ratio index between each step；
The node analysis view generation submodule of the same name shows the node of the same name for generating node analysis view of the same name
Number is reached, number is exited and exits rate, the node of the same name refers to the node for carrying out identical game behavior by player；
The loss plays furniture body Track View and generates submodule, plays furniture body Track View for generating to be lost, and shows from appointing
One node of the same name leaves the player of game specific track experienced before node of the same name herein.
9. game role behavioral data processing system as claimed in claim 8, it is characterised in that: player's footprint view is raw
Further include interactive unit at submodule, for supporting user's interaction, checks player's footprint situation of change to facilitate；The loss is played
It further includes interactive unit that furniture body Track View, which generates submodule, for supporting user's interaction, and checks the tool of each node player
The accounting of body action trail and/or each concrete behavior track.
10. game role behavioral data processing system as claimed in claim 9, it is characterised in that: player's footprint view
Generating submodule further includes that first threshold adjusts unit, shows player's footprint varied number for controlling with selectivity.
11. game role behavioral data processing system as claimed in claim 8, it is characterised in that: the node analysis of the same name
View generation submodule further includes sequencing unit, for the arrival number, exit number and exit rate and be ranked up respectively.
12. game role behavioral data processing system as claimed in claim 8, it is characterised in that: the node analysis of the same name
Duplicate removal unit is still further comprised in view generation submodule, for carrying out deduplication operation to player.
13. game role behavioral data processing system as claimed in claim 8, it is characterised in that: furniture body is played in the loss
Temporally inverted order is shown for the concrete behavior track of each node player in Track View.
14. game role behavioral data processing system as claimed in claim 9, it is characterised in that: furniture body is played in the loss
Track View, which generates, is additionally provided with second threshold adjusting unit in submodule, show the specific of different accountings for adjusting with selectivity
Action trail.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510741360.1A CN106621328B (en) | 2015-11-04 | 2015-11-04 | A kind of game role behavioral data processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510741360.1A CN106621328B (en) | 2015-11-04 | 2015-11-04 | A kind of game role behavioral data processing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106621328A CN106621328A (en) | 2017-05-10 |
| CN106621328B true CN106621328B (en) | 2019-07-26 |
Family
ID=58851257
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510741360.1A Active CN106621328B (en) | 2015-11-04 | 2015-11-04 | A kind of game role behavioral data processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106621328B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108549594B (en) * | 2018-03-30 | 2021-07-23 | 武汉斗鱼网络科技有限公司 | Method and device for determining user loss reason |
| CN108771869B (en) * | 2018-06-04 | 2022-02-08 | 腾讯科技（深圳）有限公司 | Performance test method and device, storage medium and electronic device |
| CN110852780A (en) * | 2019-10-08 | 2020-02-28 | 百度在线网络技术（北京）有限公司 | Data analysis method, device, equipment and computer storage medium |
| CN113457166A (en) * | 2021-07-20 | 2021-10-01 | 网易（杭州）网络有限公司 | Game player churn information processing method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101172202A (en) * | 2006-10-31 | 2008-05-07 | 史克威尔·艾尼克斯股份有限公司 | A network game system, a network game terminal, a displaying method and a computer program stored medium |
| CN102016745A (en) * | 2008-01-23 | 2011-04-13 | 加州大学评议会 | Systems and methods for behavioral monitoring and calibration |
| CN104731314A (en) * | 2013-12-20 | 2015-06-24 | 联想（新加坡）私人有限公司 | Providing last known browsing location cue using movement-oriented biometric data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8456298B2 (en) * | 2010-11-02 | 2013-06-04 | Timo Valtonen | Apparatus and method for portable tracking |
| US20140081715A1 (en) * | 2012-09-19 | 2014-03-20 | Salesforce.Com, Inc. | Systems and methods of coaching users in an on-demand system |
-
2015
- 2015-11-04 CN CN201510741360.1A patent/CN106621328B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101172202A (en) * | 2006-10-31 | 2008-05-07 | 史克威尔·艾尼克斯股份有限公司 | A network game system, a network game terminal, a displaying method and a computer program stored medium |
| CN102016745A (en) * | 2008-01-23 | 2011-04-13 | 加州大学评议会 | Systems and methods for behavioral monitoring and calibration |
| CN104731314A (en) * | 2013-12-20 | 2015-06-24 | 联想（新加坡）私人有限公司 | Providing last known browsing location cue using movement-oriented biometric data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106621328A (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106621328B (en) | A kind of game role behavioral data processing method and system | |
| CN108304556B (en) | The personalized recommendation method combined based on content with collaborative filtering | |
| CN107193858B (en) | Intelligent Service application platform and method towards multi-source heterogeneous data fusion | |
| Siroker et al. | A/B testing: The most powerful way to turn clicks into customers | |
| Wendy Zhu et al. | An empirical examination of guests' adoption of interactive mobile technologies in hotels: Revisiting cognitive absorption, playfulness, and security | |
| CN103544188B (en) | The user preference method for pushing of mobile Internet content and device | |
| CN107665444A (en) | A kind of web advertisement acute effect evaluation method and system based on the online behavior of user | |
| Hmedna et al. | Identifying and tracking learning styles in MOOCs: A neural networks approach | |
| CN104700152A (en) | Method for predicting tobacco sales volumes by means of fusing seasonal sales information with search behavior information | |
| CN111723973A (en) | Learning effect optimization method based on user behavior causal relationship in MOOC log data | |
| CN109214647B (en) | Method for analyzing overflow effect among online access channels based on network access log data | |
| Yang et al. | A three‐stage model of requirements elicitation for Web‐based information systems | |
| Chatzopoulos et al. | Digitization and Lean Customer Experience Management: success factors and conditions, pitfalls and failures | |
| Theusinger et al. | Analyzing the footsteps of your customers | |
| CN107122898A (en) | A kind of end-to-end SaaS air control methods of trade based on data statistics | |
| CN110069686A (en) | User behavior analysis method, apparatus, computer installation and storage medium | |
| Smith et al. | Mining controller inputs to understand gameplay | |
| CN112016000A (en) | Movie group recommendation method and system based on convolution collaborative filtering | |
| Gofman | Consumer driven multivariate landing page optimization: overview, issues and outlook | |
| CN107995289B (en) | A kind of the behaviortrace method and platform of the network user | |
| CN109345294A (en) | Wechat method for pushing and its system based on consumption habit | |
| Cebula | beyond economic and cultural capital: Network correlates of consumption tastes and practices | |
| Omar et al. | Measuring flow in gaming platforms | |
| Jadidi et al. | The hipster paradox in electronic dance music: How musicians trade mainstream success off against alternative status | |
| Xu et al. | Prediction model of music popular trend based on NNS and DM technology |
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 |