TW201120672A - Method and system for prophesying behavior of online gamer, and computer program product thereof - Google Patents

Method and system for prophesying behavior of online gamer, and computer program product thereof Download PDF

Info

Publication number
TW201120672A
TW201120672A TW098142520A TW98142520A TW201120672A TW 201120672 A TW201120672 A TW 201120672A TW 098142520 A TW098142520 A TW 098142520A TW 98142520 A TW98142520 A TW 98142520A TW 201120672 A TW201120672 A TW 201120672A
Authority
TW
Taiwan
Prior art keywords
player
game
behavior
records
online
Prior art date
Application number
TW098142520A
Other languages
Chinese (zh)
Inventor
Polly Huang
Sheng-Wei Chen
Pin-Yun Tarng
Original Assignee
Univ Nat Taiwan
Academia Sinica
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 Univ Nat Taiwan, Academia Sinica filed Critical Univ Nat Taiwan
Priority to TW098142520A priority Critical patent/TW201120672A/en
Publication of TW201120672A publication Critical patent/TW201120672A/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A63SPORTS; GAMES; AMUSEMENTS
    • A63FCARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/60Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor
    • A63F13/67Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor adaptively or by learning from player actions, e.g. skill level adjustment or by storing successful combat sequences for re-use
    • AHUMAN NECESSITIES
    • A63SPORTS; GAMES; AMUSEMENTS
    • A63FCARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/70Game security or game management aspects
    • A63F13/79Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories
    • AHUMAN NECESSITIES
    • A63SPORTS; GAMES; AMUSEMENTS
    • A63FCARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F2300/00Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
    • A63F2300/50Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
    • A63F2300/53Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers details of basic data processing
    • A63F2300/535Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers details of basic data processing for monitoring, e.g. of user parameters, terminal parameters, application parameters, network parameters
    • AHUMAN NECESSITIES
    • A63SPORTS; GAMES; AMUSEMENTS
    • A63FCARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F2300/00Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
    • A63F2300/50Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
    • A63F2300/55Details of game data or player data management
    • A63F2300/5546Details of game data or player data management using player registration data, e.g. identification, account, preferences, game history
    • AHUMAN NECESSITIES
    • A63SPORTS; GAMES; AMUSEMENTS
    • A63FCARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F2300/00Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
    • A63F2300/50Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
    • A63F2300/55Details of game data or player data management
    • A63F2300/5593Details of game data or player data management involving scheduling aspects
    • AHUMAN NECESSITIES
    • A63SPORTS; GAMES; AMUSEMENTS
    • A63FCARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F2300/00Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
    • A63F2300/60Methods for processing data by generating or executing the game program
    • A63F2300/6027Methods for processing data by generating or executing the game program using adaptive systems learning from user actions, e.g. for skill level adjustment

Abstract

A method and a system for prophesying a behavior of an online gamer, and a computer program product thereof are provided. In the present method, a gamer descriptor of a specific gamer of an online game accumulated before a time point is obtained firstly. Then, the gamer descriptor is divided into a plurality of sub-traces according to the time. Thereafter, at least one feature is derived from each of the sub-traces, and all features derived from the sub-traces are used for determining a possibility of the specific gamer to depart from the online game within a specific time period.

Description

201120672 U/A-uy 丨 003 32826twfdoc/d 六、發明說明: 【發明所屬之技術領域】 -種種㈣概,且制是有關於 產=制線上遊戲玩豕之行為的方法、·系統以及電腦程式 【先前技術】 得」rim的晋及化,現代人逐漸習慣透過網路取 人聯繫,也有越來越多的現代人會以上網作 為主要的休閒活動,_帶動了線 =統單顧,是,線上遊戲增加了玩家二: _ 66礙玩^們此自仃組成公會進而集體活動、彼此較勁, 生*練欠化私度與新鮮感更因此大幅提升。根據統計,近 年來、.泉上遊戲的市場仍在與日俱增當中。 ^般來說,線上遊戲的收費制度包括月費制、時數 :一以,虛㈣物販賣制。但無論是採取何種收費制度, 多寡都將對遊戲業者的利潤造成最直接的影 曰。寸別是願意花費大量時間與金錢投入遊戲的重度玩家 anl_coi*e piayer) ’對線上遊戲業者的獲利影響甚鉅。 ,正口如,,如何追蹤每位玩家的對遊戲的觀感及忠誠 又Ϊ疑疋線上遊戲業者欲提升產值所必須關注的議題。 、j而,目说有關於線上遊戲之玩家行為的研究多半是 以—群玩豕的行為做為分析單位。例如探討同一伺服器的 人數變化狀況’或是網路穩定與否對玩家人數的影響 等等這些研究大多是用以推算在將來玩家人數的增減情 201120672, v / w〆 * vv) 32826twf.doc/d 況’但不足以預測出哪些玩家可能會長期地退出遊戲,因 此能提供給線上遊戲業者的資訊仍有所不足。 【發明内容】 有鑑於此,本發明提供一種線上遊戲玩家行為之預測 方法,以玩家為單位預測在未來一段期間内該名玩家是否 會長期地離開線上遊戲。 ^發明提供—種線上遊戲玩家行為之預啦統,能預 測出單一玩家準備離開線上遊戲的行為。 ^發明提供-種電腦程式產品,所儲存的程式指令在 统後’可使電腦系統具備預測線上遊戲之玩家 在一日提出—種線上遊戲玩家行為之預測方法,用以 在日寸間點預測線上遊戲之特定玩家的行為。此 ==日㈣點之前所累積關於蚊玩家 記 ===描述記錄劃分為多段子記錄,並針夂 錄所萃取出的特徵值, 曼,據伙口子把 離開線上遊戲的可能性。疋玩豕在一未來指定期間内 在本發明之一實施例中,農 特定玩家的遊戲相關記錄以及特,錄至少包括 中之一。 寻疋玩豕的生活相關記錄其 在本發明之一實施例中, 上下='角_,角色包括 在本發明之-實施例中,其中當玩家描述記錄包括上 201120672 w^l〇〇3 32826twf.doc/d 包括根^二錄萃取出至少―特徵值的步驟 間平均值、投記=,每_時 時間平均佶 ^八 句線盼間點、每次上線 作母曰it戲時間變異數至少其中之-,以 作為子S己錄所對應的特徵值。 ^發明之—實施财,射生活 玩豕貢料以及玩家舉止表現其中之_。 丁 ^匕栝 在本發明之-實施例中,其中依據從各子 : = = j斷特定玩家在未來指定期間内離開線上遊 =可此性的步驟包括利用機器學習機制處理從各子 2取出的特徵值,以計算特定玩家在未來指定期間内離 幵、、友上遊戲的可紐。其t,機器學習機制可以是監督 ^^rViSed)學習分類法或非監督式(職-supervised)學 二为類法,*m學習分祕包括讀向錢201120672 U/A-uy 丨003 32826twfdoc/d VI. Description of the invention: [Technical field of invention] - Various methods (4), and the system, method, and computer program for the behavior of online game play [Prior Art] With the promotion of "rim", modern people are gradually accustomed to getting in touch through the Internet, and more and more modern people will use the Internet as their main leisure activity, _ driving the line = unified single, is Online game has increased the number of players: _ 66 hinder the play ^ These self-proclaimed guilds and then collective activities, each other's strength, the students * practice less private and fresh sense, so significantly improved. According to statistics, in recent years, the market for games on the spring is still growing. Generally speaking, the online game charging system includes the monthly fee system and the number of hours: one, the virtual (four) goods selling system. However, no matter what kind of charging system is adopted, the amount will have the most direct impact on the profit of the game operators. Inch is a heavy player who is willing to spend a lot of time and money on the game. anl_coi*e piayer) 'The profitability of online gamers is huge. , the mouth is like, how to track each player's perception and loyalty to the game and doubt the issues that online game operators must pay attention to in order to increase production value. J, and the study of the behavior of players in online games is mostly based on the behavior of group play. For example, to discuss the change in the number of people on the same server or the impact of the stability of the network on the number of players, etc. These studies are mostly used to estimate the increase or decrease of the number of players in the future 201120672, v / w〆* vv) 32826twf. Doc/d condition 'but not enough to predict which players may quit the game for a long time, so the information available to online gamers is still insufficient. SUMMARY OF THE INVENTION In view of the above, the present invention provides a method for predicting the behavior of an online game player, predicting whether the player will leave the online game for a long period of time in the future. The invention provides a pre-event of online game player behavior that predicts the behavior of a single player preparing to leave the online game. ^Invention provides a kind of computer program product, the stored program instructions in the post-system can make the computer system have the player who predicts the online game to be proposed in one day - a prediction method of online game player behavior, used to predict the time between days The behavior of specific players in online games. This == Day (4) points accumulated before the mosquito player record === Description record is divided into multi-segment sub-records, and the feature values extracted by the record, Man, the possibility of leaving the online game.疋 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕 豕Looking for a life-related record of play, in one embodiment of the present invention, up and down = 'corner_, the character is included in the embodiment of the present invention, wherein when the player description record includes the above 201120672 w^l〇〇3 32826twf .doc/d includes the average value of the steps of extracting at least the eigenvalues from the roots and the second record, and the average of 时间^8 sentences per line, and the time difference between each time. At least one of them is used as the feature value corresponding to the child S record. ^Invented - Implementing money, shooting life, playing tribute and player behavior. In the embodiment of the present invention, wherein the step of leaving the line upstream within a specified period of time from the respective sub-sections: ==j = the step of including the use of a machine learning mechanism to process the removal from each sub- 2 The feature value is used to calculate the specific player's ability to leave the game during the specified period of time in the future. Its t, the machine learning mechanism can be supervised ^^rViSed) learning taxonomy or unsupervised (supervised) learning two class method, *m learning sub-secret including reading money

Vector Machine,SVM)。 PP 從另-觀點來看’本發明提出—種線上遊戲玩家 之預測系統’此系統包括輸人/輸.出介面、儲存單元 值萃取單元以及預測單元。其中,儲存單元用以儲存線^ 遊戲之多個玩家個觸玩家描述記錄。舰值萃取單 、 接輪入/輸出介面與儲存單元,在輪入/輸出介面取得特 玩家與一時間點後,特徵值萃取單元自儲存單元取得在萨 間點之前所累積關於特定玩家的玩家描述記錄,並依I 將玩家描述記錄劃分為多段子記錄,以及針對每個子圮^ 萃取出至少一特徵值。預測單元耦接輸入/輸出介面與特= 201120672 07A-091003 32826twf.doc/d ^ ^依據從各子記錄所萃取出的特徵值來判 豕在未來指錢間内_線上遊戲的可能性,並 透過輸入/輪出介面輸出可能性。 斤―2發明之—實施例巾’其巾玩家描述記錄至少包括 =一_遊齡關記錄以及特定玩家的生活相關記錄其 卜,本么月之Λ知例中’其中遊戲相關記錄至少包括 上下線記錄、角色資料’以及肖色舉止表現其中之一。 在本發明之一實施例中,盆中 __ ^ 描述記錄包括上下線⑽味4Ί綠卒取早几在玩豕 所八龍泉錄時,根據各子記錄推算各子記錄 斤刀別對應的母日遊戲時間平均值、投人遊 均上線時間點、每次上線砗 八 千 變_值、以及每日遊戲時間 ^ ^ ,以作為子記錄所對應的特徵值。 玩明之—貫施例中’其巾生活相關記錄至少包括 玩豕貧料以及玩家舉止表現其中之一。 羽機ϊίίΓ之一實施例中,其中預測單元係利用機器學 二太去:各子記錄所萃取出的特徵值,以計算特定玩 ίk躺嶋開線上遊戲的可能性。其中,機器 ζ ”制包括監督式學習分類法以及非監督式學習分 法。而監督式學習錢法包括續向量機。 觀點來看,本發明提出一種電腦程式產品,所 個程式指令在载人電腦系統並執行之後,可執行 上述線上遊戲玩家行為之預測方法。 基於上述’本發明係取得線上遊戲之單一特定玩家的 201120672 ^^-^>1003 32826twf.doc/d 述,,並從中分析萃取出多個特徵值,以利用上 晴名特定縣是否會在未來—段期間内長 開此線上,。如此-來,便有機會在玩家對遊戲 ^王喪失熱忱之則,改進遊戲内容以提升玩家繼續進行遊 戲的意願。 —為讓本發明之上述特徵和優點能更明顯易懂,下文特 舉實施例,並配合所附圖式作詳細說明如下。Vector Machine, SVM). PP From another point of view, the present invention proposes a prediction system for online game players. This system includes an input/output interface, a storage unit value extraction unit, and a prediction unit. Wherein, the storage unit is used to store the plurality of players of the game, and the player describes the record. The ship value extraction single, the wheel input/output interface and the storage unit, after the special input player and the time point are obtained in the wheel input/output interface, the feature value extraction unit obtains the player who has accumulated the player before the S-point from the storage unit. Describe the record, and divide the player description record into multiple sub-records according to I, and extract at least one feature value for each sub-^. The prediction unit is coupled to the input/output interface and the special = 201120672 07A-091003 32826twf.doc/d ^ ^ based on the eigenvalues extracted from each sub-record to determine the possibility of _ online game in the future reference money, and The possibility of output is output through the input/rounding interface.斤 - 2 invention - the embodiment towel 'the towel player description record at least includes = one _ 游 游 off record and the specific player's life related record, in this month's Λ Λ ', where the game related records at least include the upper and lower lines One of the records, the character data', and the black behavior. In an embodiment of the present invention, the __^ description record in the basin includes the upper and lower lines (10), the taste of the green, and the green strokes are recorded in the early days of the play, and the corresponding records are calculated according to the respective sub-records. The average value of the daily game time, the time when the investment is on the line, the time of the online change, the value of the daily game time ^ ^, and the feature value corresponding to the child record. In the case of play--the example of the life of the towel, at least one of the performances of the game and the performance of the player's behavior. In one embodiment, the predictive unit utilizes the machine to learn the feature values extracted by each sub-record to calculate the likelihood that the particular game will lie on the online game. Among them, the machine system includes a supervised learning classification method and an unsupervised learning method. The supervised learning money method includes a continuous vector machine. From the viewpoint, the present invention proposes a computer program product, and the program instructions are in manned After the computer system is executed, the above-mentioned online game player behavior prediction method can be executed. Based on the above description, the present invention is a single specific player who obtains the online game, 201120672 ^^-^>1003 32826twf.doc/d, and analyzes therefrom. Extracting a number of eigenvalues to take advantage of whether the specific county in the Qingming name will open the line in the future period. If so, there will be an opportunity to improve the game content after the player loses enthusiasm for the game. The willingness of the player to continue the game is enhanced. - The above-described features and advantages of the present invention will become more apparent and understood from the following detailed description.

【實施方式】 由於線上遊戲之玩家對遊戲的觀感會反應在該名玩 ^的=_錄當中’因此絲透過相關記錄的分析而推測 一玩豕的行為,便能在玩家的滿意度低落時對遊戲内容進 行改進’以健玩家對遊戲的銳。本發日績是基於上述 觀點進而發展出的一種線上遊戲玩家行為之預測方法、系 統與電腦程式產品。為了使本發明之内容更為明瞭,以下 特舉實施例作為本發明確實能夠據以實施的範例。 圖1是依照本發明之-實施例所緣示之線上遊戲玩家 行為之預測系統的方塊圖。請參閱圖i,線上遊戲玩家行 為之預測系統100包括輸入/輸出介面11〇;儲存單元12〇、 特徵值萃取單元130,以及預測單元140。線上遊戲玩家行· 為之預測系統10 0係用以預測一線上遊戲之單—玩家的行 為’特別是針對玩家不再遊玩此線上遊戲的行為進行預測。 在本實施例中,輸入/輸出介面110例如是鍵盤與螢幕 的組合,或者是觸控式螢幕等等,用以接收預測目標的輸 入並且輸出預測結果。儲存單元120例如是記憶體、記憶 201120672 07A-091003 32826twf.doc/d 卡或硬料任何f轉單元,在 例中儲存單元m所儲存的是不斷=線=本實施 玩家個別的玩家描述記錄。 累積的線上遊戲之各個 特徵值萃取單元130耦接 單元120,用以在輪入/輪出^輪;"/輸出介面110與儲存· 特定玩家後,自儲存單元接㈣為預測目標的 並從中萃取出多個:取传相關的玩家描述記錄, 單元^單:140耦接至輸入’輸出介面110與特徵值萃取 =判=據特徵值萃取單元13〇所萃取出的特: ⑽。_動作’並將判斷的結果顯示於輸人/輸出介面 _ ϋ更進―、步地說m稍玩家行為之預測系統 〜、B。平、、田運作在程’以下特舉另一實施例來對本發明進 圖2疋依照本發明之—實施例所繪示之線上遊戲 豕行為之預測方法的流程圖,請同時參關丨與圖2。 _—在透過輸入/輸出介面110取得作為預測目標的特定 ί家j及預測當時的時間點後,如步驟210所示,特徵值 卒取單,m自儲存單元uo轉在此賴點之前所累積 關於=定玩_玩家描述麟。在本實關巾,特定玩家 的玩豕g述記錄至少包括特定玩家的遊齡關記錄以及生 ,相關g己錄其中之一’但並不以此為限。具體來說,遊戲 錄至少包括上下線記錄、角色資料,以及角色舉止 又現其中之—。而生活相關記錄則至少包括玩家資料以及 玩家舉止表現其中之一。 201120672 U/A-uyl003 32826twf.doc/d …接下來在步驟220 t,特徵值萃取單元130依時間將 ,豕^5&_分為多段子記錄。舉例而言,特徵值萃取 ^ W記錄等分為κ段子記錄。其 ’ 以疋10或其他正整數,在此同樣不加以限制。 子驟23G所示,特徵值萃取單元13G針對每個 =錄,出-個或—個以上的特徵值。必須說明的是, 岐錄之内容的不同’特徵值萃取單元130所 錄勺舰—徵Ϊ種類也不相同。舉例來說,#玩家描述記 疋玩豕的上下線記錄時’特徵值萃取單元130可 記Γ内容來推算每曰遊戲時間平均值'投入遊 」、’、又平均上線時間點、每次上線時間平均值、以及 t日遊戲_變異數至少其中之―,進而作為各子記錄所 分別對應的特徵值。 “ 而在玩家描述記錄包括角色資料時,特徵值萃取草元 貝I:以特疋玩家在線上遊戲中所扮演之角色的種族、 速度或裝備等資料來作為特徵值。當玩家描述 4°士定、二所f舉止表現時,特徵偉萃取單元130則可依據 ^疋玩豕所扮演之角色在遊戲中的消費狀況,或與盆他角 ==互動_來取得特徵值。此外,在玩家描述記錄 ^括玩豕貧料時,特徵值萃取單幻如將以玩家本身 2 (例如性別、學歷、職業、居住地,或收人等等)作為 ,徵^而在玩家描述記錄包括玩家舉止表現時,特徵值 130則會分析玩家在現實生活令的實際行為(例 如付秋方式或付款延遲與否等)來取得特徵值。 201120672 32826twf.doc/d 必須特別說明的是,上述玩家描述記錄以及特徵值的 麵僅是本發明的—些實施卜線上補玩家行為之預 測系統100在進行預測時,並非僅能以上述玩家描述記錄 以及特徵值來作為判斷依據。進_步來說,任何與玩 關,靜態或動態資訊均可成為玩家描述記錄,而特徵值萃 取單元13G會根據不同的玩家描述記錄萃取丨不同的特徵 值。 所-ίϊ如步驟所示,預測單7^ 140依據從各子記錄 ^ f的所树難,_歡玩家在某—未來指定期 i戲的;t來十天内、或未來一個月内等等)離開線上 在本實施例中,預測單元140係利用-機 制f里從各子記錄所萃取出的所有特徵值,進而 豕在未來指定顧_開線上遊戲的可能性。 二非?何以是監督式(super— 習八‘彳ΐ lnGn_supen/ised)學習分類法。而監督式學 ?、疋支持向量機(Support Vector Machine, /· 士 對特徵值進行分類以取得預測結果。 在去W二,例中,預測單元14G可判斷特定玩家是否會 未開線上遊戲,或者是計算特定玩家在 機率™結果可透 系對上遊戲玩家行為之預測 在任何的時間點,根據目if:個玩家進行追縱,從而 康目則為止累積蒐集到的玩家描述記 201120672 1003 32826twf.doc/d 錄,預測未來一段時間内某位玩家是否會出現最後—次上 線的行為。一旦預測出玩家即將會長期地停止遊玩此線上 遊戲,便可以立即進行調查,從而積極地對遊戲内容進 改善,崎低玩家㈣失率。另外,線上稍業者也能根 據線上遊戲玩家行為之預測系統1〇〇的判斷結果,對可护 停止遊玩的玩家作進-步的分析,判斷這些即將離開遊^[Embodiment] Since the player's perception of the game in the online game will be reflected in the =_record of the name of the game, it is possible to speculate on the behavior of the game through the analysis of the relevant record, and the player's satisfaction can be lowered. Improve the game content's sharpness to the game. This is a prediction method, system and computer program product for online game player behavior based on the above viewpoint. In order to clarify the content of the present invention, the following specific examples are given as examples in which the present invention can be implemented. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram of a prediction system for online game player behavior in accordance with an embodiment of the present invention. Referring to Figure i, the online game player behavior prediction system 100 includes an input/output interface 11A; a storage unit 12A, an eigenvalue extraction unit 130, and a prediction unit 140. Online game player line · For the prediction system, the system is used to predict the behavior of the online game - the behavior of the player', especially for the behavior of the player no longer playing this online game. In the present embodiment, the input/output interface 110 is, for example, a combination of a keyboard and a screen, or a touch screen or the like for receiving an input of a predicted target and outputting a predicted result. The storage unit 120 is, for example, a memory, memory 201120672 07A-091003 32826twf.doc/d card or any f-transfer unit of the hard material. In the example, the storage unit m stores the constant = line = the player's individual player description record. Each feature value extraction unit 130 of the accumulated online game is coupled to the unit 120 for use in the wheeling/rounding of the wheel; "/output interface 110 and storing the specific player, and then connecting (4) from the storage unit to the predicted target Extracting a plurality of them: taking the relevant player description record, unit unit: 140 is coupled to the input 'output interface 110 and feature value extraction=judgement= according to the characteristic value extraction unit 13〇 extracted: (10). _Action' and display the result of the judgment on the input/output interface _ ϋ 进 ― 、 、, step by step m predictive system of player behavior ~, B. The following is a flowchart of the method for predicting the online game behavior of the present invention according to the embodiment of the present invention, and please refer to the following. figure 2. _—After obtaining the specific γ home as the prediction target through the input/output interface 110 and predicting the time point at that time, as shown in step 210, the eigenvalue ticket is taken, and m is transferred from the storage unit uo before the lag point. Cumulative about = fixed play _ player description Lin. In the actual customs towel, the specific player's play record includes at least the specific player's age record and the student's record, but not limited to this. Specifically, the game record includes at least the upper and lower line records, character data, and character behaviors. Life-related records include at least one of player profiles and player behavior. 201120672 U/A-uyl003 32826twf.doc/d ... Next at step 220 t, the feature value extraction unit 130 divides 豕^5&_ into a plurality of sub-records according to time. For example, the eigenvalue extraction ^ W record is equally divided into κ segment sub-records. Its ' is 10 or other positive integers, and is also not limited here. As shown in sub-step 23G, the feature value extracting unit 13G records, for each =, one or more feature values. It must be noted that the difference in the content of the transcript is different from the type of the sampan-type of the eigenvalue extraction unit 130. For example, when the player describes the upper and lower line records of the game play, the feature value extraction unit 130 can record the content to calculate the average value of each game time 'investment tour', 'and the average online time point, and each time online. The time average value and the t-day game_variation number are at least one of them, and further, the feature values corresponding to the respective sub-records. “When the player description record includes the character data, the feature value extracts the grass dollar I: as the feature value of the character, speed or equipment of the character played by the player on the online game. When the player describes 4° When the performance of the first and second performances is performed, the feature extraction unit 130 can obtain the feature value according to the consumption status of the role played by the game, or interact with the pool angle == to obtain the feature value. When the description record is included in the game, the feature value extraction single illusion will be based on the player itself 2 (such as gender, education, occupation, place of residence, or income, etc.), and the player description record includes the player's behavior. In performance, the feature value 130 analyzes the player's actual behavior in real life orders (such as paying for the fall mode or delay in payment, etc.) to obtain the feature value. 201120672 32826twf.doc/d It must be specially stated that the above player description record And the face of the feature value is only the present invention - the predictive system 100 of the player player behavior is not only able to use the above player description record and the feature value. For the purpose of judging. In any step, the static or dynamic information can be the player description record, and the feature value extraction unit 13G will extract different feature values according to different player description records. As shown in the step, the forecasting list 7^140 is based on the difficulty of the tree from each sub-record, and the player is leaving the line in a certain future period of the specified period; t to ten days, or the next month, etc. In this embodiment, the prediction unit 140 utilizes all the feature values extracted from the respective sub-records in the mechanism f, and then specifies the possibility of playing the game on the open line in the future. Xi Ba '彳ΐ lnGn_supen/ised) learning classification method, and supervised learning?, Support Vector Machine (Support Vector Machine, /· classification of eigenvalues to obtain prediction results. In W2, in the case, prediction The unit 14G can determine whether the specific player will not play the online game, or calculate the probability that the specific player is in the probability TM can be compared to the behavior of the upper game player at any point in time, according to the target if: a player to track, Therefore, Kang Mu has accumulated the collected player descriptions 201120672 1003 32826twf.doc/d, predicting whether a player will have the last-on-line behavior in the future. Once the player is predicted, the player will stop playing for a long time. Online game, you can immediately investigate, so as to actively improve the game content, low player (four) failure rate. In addition, the online industry can also be based on the online game player behavior prediction system 1 〇〇 judgment results, can be protected Players who stop playing play an analysis of the step-by-step, judging that these are about to leave the tour ^

特定的飼服器。若是,則可能是因為 連^顧導致玩豕喪失熱忱。據此,對_器進行_ 檢查以就硬體或網路設備層面做出改進Q ^3S依照本發明之另—實施例所緣示之線上 ^丁為之賴方法的流程圖。在以下的實施财,是= 下線記錄來作為玩家描述記錄。請參閱圖3,在任 务準備對—線上遊戲的—特定玩家進行預測時,首 ’取得在此時間點之前所累積關於此特 錄。其巾’上下線記錄包括玩家從開妗 遊戲-直到進行預測當時,每天的上線及二 接著在步驟320中,將特定玩家的上下 劃分(例如等分)為多段子記錄 ^ =間 預測時特技家已遊玩此線上遊戲⑽^兄:::進行 分為—段錄,將第34至66天劃 接下:==二劃二^ 母日遊戲和平均贿投人度來作^記二^ 201120672 υ/Α-υ^ιυυ3 32826twf.doc/d 應的特徵值。具體來說,每日遊戲時間平均值係表示 記錄所涵蓋的時間内,特定玩家上線時間總和與子記錄所 ,蓋之天數的比值。而投人遊戲密集度則是在子記錄所 蓋的時間内,特定玩家有上線之天數與子記錄所涵蓋之天 數的比值。 在計算出每個子記錄所對應的每曰遊戲時間平均 以及投入遊戲密集度後,最後在步驟34〇中,依據各子記 錄=對應的每日遊戲時間平均鋪投人_密#度,判斷 特定玩家在未來指定期間内離開線上遊戲的可能性。在本 實施例中’例如是將計算出的所有每日遊戲時間平均值盘 投入遊戲錢度輸人已事先經支持向量機來進行 分類處理’從而預珊定玩家是否會在未來指定期間内離 開線上遊戲。 —在本實關巾,為了增加制精確度,亦可加入特定 玩家的其他種玩家描述記錄(例如玩家資料、玩家舉止表 現、角色資料,或角色舉止表現料)來萃取出更多的特Specific feeding device. If so, it may be because of the loss of enthusiasm caused by the game. Accordingly, a _ check is performed on the ___ to perform a modification on the hardware or network device level. Q ^ 3S is a flow chart of the method according to another embodiment of the present invention. In the following implementation, the = offline record is used as the player description record. Referring to Figure 3, when the task is to prepare a prediction for a particular player of the online game, the first 'acquisition is accumulated before this point in time. The towel's upper and lower line records include the player from the opening game - until the prediction is made, the daily go-on and the second then in step 320, the specific player's up and down division (eg, halving) into multiple sub-records ^ = inter prediction The family has already played this online game (10) ^ brother::: is divided into paragraphs, recorded on the 34th to 66th day: == two strokes ^ ^ mother day game and average bribe investment to make ^ ^ 2 201120672 υ/Α-υ^ιυυ3 32826twf.doc/d The characteristic value should be. Specifically, the daily game time average is the ratio of the total time of the specific player's online time to the number of days covered by the record, and the number of days covered. The intensive game play is the ratio of the number of days a particular player has lived to the number of days covered by the sub-record during the time covered by the sub-record. After calculating the average game time and the game intensity corresponding to each sub-record, finally, in step 34, determining the specificity according to each sub-record=the corresponding daily game time average spreader_密# degree The possibility that the player will leave the online game for a specified period of time in the future. In the present embodiment, for example, the calculated average value of all the daily game time is put into the game money. The input has been previously processed by the support vector machine to determine whether the player will leave in the specified period in the future. Online game. - In the actual customs towel, in order to increase the accuracy of the system, other player description records (such as player data, player behavior, character data, or character behavior) can be added to extract more special features.

徵值。接著再把所有取得的特徵錄人讀向量機以取得 預測結果。 本發明另提供一種電腦程式產品’其係用以執行上述 線上遊戲玩家行為之預測方法。此電腦程式產品基本上是 由數個程式指令片段所組成(例如設定程式指令片段、以 =署程式指令片段料)’在將這些程式指令片段載入 電腦系統並騎之後,即可完成上述線上賴玩家行為之 預測方法之各步驟,並使得電腦系缝藉由分析單一玩家 12 201120672 υ/«-υ?1003 32826twf.doc/d 的玩家描述記錄,以預測兮文 a 線上遊戲的可能性。〜 豕在未來某段時間内離開 綜上所述,本發明所述之線上 法、系統及電腦程式產品係針對二二丁 ',·、預測方 測。據此,線上遊戲?者能在、條豕物為做預 续卜游_ 4 在預測出玩家即將長期離開此 敎㈣,透過問卷調錢以找出造成玩家喪失遊戲 有、^原因,從而對遊戲作出適當的改進。此外 Γ_戲的玩家,線上遊戲業者也能對其 =體層面與硬體層面❹方改進,達到鞏固玩家數量的目 雖然本购已以實施_露如上,财並非用以限定 ^明’任何所屬技術領域中具有通f 神和範圍内,當可作些許之更動與潤=: 二明之保€範圍當視後附之巾請專利範_界定者為準。 I圖式簡單說明】 一 图1疋依知、本發明之一實施例所纟會示之線上遊戲玩家 仃為之預測系統的方塊圖。 〜圖2是依照本發明之一實施例所繪示之線上遊戲玩家 订為之預測方法的流程圖。 ^ 圖3是依照本發明之另一實施例所繪示之線上遊戲玩 豕行為之預測方法的流程圖。 【主要元件符號說明】 1(30 :線上遊戲玩家行為之預測系統 201120672 v / i w3 ^2826twf.doc/d no:輸入/輸出介面 120 :儲存單元 U0 :特徵值萃取單元 140 :預測單元 210〜240 :本發明之一 為之預測方法的各步驟 實施例所述之線上遊戲玩家行 / 310〜340 :本發明之另一實施例所述之線上遊戲玩家 行為之預測方法的各步驟Value. Then, all the acquired features are recorded in the vector machine to obtain the predicted result. The present invention further provides a computer program product 'which is used to perform the above-described prediction method of online game player behavior. This computer program product basically consists of several program instruction fragments (such as setting a program instruction fragment, and using the program instruction fragment). After loading these program instruction fragments into the computer system and riding, the above line can be completed. Relying on the steps of the player's behavior prediction method, and making the computer system stitching by analyzing the player description record of a single player 12 201120672 υ/«-υ?1003 32826twf.doc/d to predict the possibility of online games. ~ 离开 Leaving for a certain period of time in the future In summary, the online method, system and computer program product described in the present invention are directed to the second and second, ', and prediction models. According to this, online games? In the event that the player can make a pre-renewed tour _ 4 In predicting that the player is about to leave this game for a long time (4), transfer money through the questionnaire to find out the cause of the player losing the game, and to make appropriate improvements to the game. In addition, the players of the game, the online game industry, can also improve their physical and hardware levels, and achieve the goal of consolidating the number of players. Although the purchase has been implemented, the above is not disclosed. In the technical field, there is a scope of the gods and the scope, when a little change and run can be made =: The scope of the protection of the two Mings is subject to the scope of the patent. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram of a prediction system which is shown by an online game player according to an embodiment of the present invention. 2 is a flow chart of a method for predicting an online game player in accordance with an embodiment of the present invention. FIG. 3 is a flow chart showing a method for predicting online game play behavior according to another embodiment of the present invention. [Main component symbol description] 1 (30: Online game player behavior prediction system 201120672 v / i w3 ^2826twf.doc/d no: input/output interface 120: storage unit U0: feature value extraction unit 140: prediction unit 210~ 240: One of the steps of the present invention is an online game player line/310~340 as described in each step embodiment of the prediction method: steps of a method for predicting the behavior of an online game player as described in another embodiment of the present invention

1414

Claims (1)

201120672 ^1003 32826twf.doc/d 七、申謗專利範園: 點預、i —種線上遊戲玩家行為之預測方法,適於在一時間 2預娜-線上賴之—特定玩家的行為,财法包括· 家描以該時間點之前所累積關於該特定玩家的-玩 依時間劃分該玩家描述記錄為多段 及針對該些子記錄的每一個,萃取出至卜、特徵值;以 斷讀特些:記錄所萃取出的該至少-特徵值,列 能性。在—未來指定期間内離開該線上遊戲的-f 預心申=範圍第1項所述之線上物晰為之 該特定玩家的述記錄至少包括下列其中之-: 相關記錄。♦戲相關§己錄,以及該特定玩家的-生活 鲁 預剛方法申it乾圍第2項所述之線上遊戲玩家行為之 〜上下綠纪;°亥:戲相關記錄至少包括下列其中之,: 預剛方法,3項所述之線上遊戲玩家行為 針對該些子命棘田,玩豕插述記錄包括該上下線記錄時 包扭. ;;的每—個,萃取出該至少一特徵值的少驊 4·如申請專利$第:以及-角色舉止表現。 、方法,其中者=3項所述之線上遊戲玩家行為之 根據各該也早^ 、每日遊戲時f推算各該好記賴分別對應的 句值、一投入遊戲密集度、一爭均上線 15 201120672 u//\-uyiw3 32826iwf.doc/d 次上線時間平均值、以及-每曰遊戲時間變 /、數至八、/之―’以作為該至少—特徵值。 ㈣申ί專利範圍第2項所述之線上遊戲玩家行為之 中該生活_記錄至少包括下列其中之-: 一玩豕貧料以及一玩家舉止表現。 預測3申ϊί利範圍第1項所述之線上遊戲玩家行為之 特徵值/ #㈣f據從各該些子記錄所萃取出的該至少一 特徵值,_該歡玩家在該未來 遊戲之該可能性的步驟包括: ㈣㈣開該線上 利用-機H學f機魏理從各 的該至少-特徵值,以叶管—于°己錄所卒取出 内離關呼綠十π該4寸疋玩豕在該未來指定期間 円離開δ亥線上遊戲的該可能性。 預範圍第6項所述之線上遊戲玩家行為之 制方法,其中該機器學習機制包括下列1 4 督式(supervised )學習分類法以及、: (non-supervised)學習分類法。 F皿7T式 8. 如申請專利範圍第7項所述之線 預測方法’其中該監督式學習讀领=二= (SupportVeotorMachine^SVM) 〇 枯支持向夏機 9. -種線上遊戲玩家行為之預測系統 一輸入/輸出介面; · 一儲存單元,儲存一線上遊戲之多 家描述記錄; 夕個玩豕個別的一玩 一特徵值萃取單元,耦接該輸入/輪 询出;丨面與該儲存單 16 201120672 -.....1003 32826twf.doc/d 兀’在該輸入/輸出介面取得 特徵值萃取單元自該儲存單 一蚪間點後,該 關於該特定玩家的該玩家點之前所累積 述記錄為多段&盼間劃分該玩家描 出至少及針對該些子記錄的每-個,萃取 _ —預測單7t,_該輸人/輸出介 嵌 各該些子記錄所萃取出的該至二:π 能性,並透财於f/tt 職線鸟戲的一可 並透過該輪入/輸出介面輸出該可能性。 之預測系Π丨二销H遊戲玩家行為 生活相遊戲相_,以及該特定玩家的- 之預i;.r:請ίΓ1圍第10項所述之線场戲玩家行為 一⑽、統’其巾該遊戲相敎駐少包括下列其中之 • 記錄、—角色#料,以及―角色舉止表現。 範圍第11項所述之線上遊戲玩家行為 徵值萃取單元在該玩家描述記錄包 所:別,根據各該些子記錄推算各該些子記錄 τ應的-母日遊戲時間平均值、—投人遊戲密集 ς平均上線時間點、—每次上線時間平均值、以及— Ζ遊戲時間變異數至少其中之一,以作為該至少一特徵 13.如申請專利範圍第1〇項所述之線上遊戲玩家行為 17 201120672 v / j\~\jy iw3 32826twf.doc/d !:賴^If中該生活相關記錄至少包括下列其中之 一.一玩豕貧料以及一玩家舉止 14·如申請專利範圍第9項^ 之預測系統,其之線上遊戲玩家行為 從各該些子記錄辭取利用—顧學習機制處理 的。Λ至少—特徵值,以計算該特 疋玩豕在該未來&定賴 之預π^Γ14項所述之線=戲二 之預m.其巾該難學f 一 監督分類法以及—非監督式學習之. 定玩遊戲之一特 依時間劃分該玩家描述記 及針對該些子記錄的每—個,萃取出=錄特徵值;以 斷該:萃取出的該至少-特徵值,到 能性。 ^㈣_該線上遊戲的4 -遊戲相關記錄,以及該特定玩家的一生活家’ 18 lu03 32826twf.d〇c/d 201120672 中,:二申:ί專利範圍第ls項所述之電腦程式產品,1 中該遊戲相關記錄至少包括下列其中之—: :、 錄、-角色資料H角色舉止表現。 I己 20.如申請專利範圍第19項所述之電腦 ^當該玩家描軌錄包減上下粒料,鱗程 ^ =該些子記錄的每—個萃取出該至少—特徵值時二 =錄推算各該些子記錄所分別對應的-每曰遊 戲ί間千均值、—投人遊戲密集度一平均上線時間點、 =次上線時間平均值、以及—每日遊戲時間變異數至 之一,以作為該至少一特徵值。 21,如申請專利範圍第18項所述之電 :該生活相關記錄至少包括下列其…:一料: 及一玩家舉止表現。 22. 如申請專利範圍第1?項所述之電腦程式產品,其 —°玄些程式指令在依據從各該些子記錄所萃取出的該至少 ==徵值’判斷雌定玩家在該未來指定_内離開該ς 蠖戲之該可能性時,利用一機器學習機制處理從各該些 2錄所萃取出的該至少一特徵值,以計算該特定玩家在 χ來指定期間内離開該線上遊戲的該可能性。 23. 如申請專利範圍第22項所述之電腦程式產品,其 碡機器學習機制包括下列其中之一:一監督式學習分类員 、以及一非監督式學習分類法。 24. 如申請專利範圍第23項所述之電腦程式產品,其 該t督式學習分類法包括一支持向量機。 19201120672 ^1003 32826twf.doc/d VII. Shenyi Patent Fanyuan: Point pre-, i---the prediction method of online game player behavior, suitable for one time 2 pre-na-line--specific player behavior, financial law Including: the homework is divided by the time-point-accumulated player-specific time-divided player description record as a plurality of segments and for each of the child records, extracting the feature value, and extracting the feature value; : Recording the extracted at least-feature value, column energy. -f The intention to leave the online game during the specified period of the future. The scope of the specific item described in item 1 is at least the following: - Related records. ♦ related § recorded, and the specific player's - life Lu pre--------------------------------------------------------------------------------------------------------------------------------------------------------- : The pre-fighting method, the online game player behavior described in the three items is for the sub-skins, and the play-intercept record includes the at least one feature value for each of the on-line records. The ensign of 4: such as the patent: $: and - the performance of the role. The method, the in-person = 3 items of the online game player's behavior according to each of the early ^, the daily game f, the calculation of each of the good deeds corresponding to the sentence value, a game intensive, a fight on the line 15 201120672 u//\-uyiw3 32826iwf.doc/d The average time of the online time, and - every game time change /, the number to eight, / ― ' as the at least - feature value. (4) The online game player behavior described in item 2 of the patent scope of the patent, the life record _ at least includes the following ones: one play poor and one player behavior. Predicting the value of the online game player's behavior as described in item 1 of the claim 3/#(4)f according to the at least one feature value extracted from each of the sub-records, _ the player’s possibility in the future game The sexual steps include: (4) (4) Open the online use - machine H learning f machine Wei Li from each of the at least - characteristic value, to the leaf tube - in the ° recorded by the death of the departure from the green 10 π the 4 inch 疋 play该The possibility of leaving the game on the δhai line during the specified period of the future. The method for determining the behavior of an online game player as described in item 6 of the pre-scope, wherein the machine learning mechanism comprises the following 14 supervised learning taxonomy and: (non-supervised) learning taxonomy. F dish 7T type 8. As described in the scope of claim 7 of the line prediction method 'where the supervised learning read collar = two = (SupportVeotorMachine ^ SVM) support for the summer machine 9. - online game player behavior Predicting an input/output interface of the system; · a storage unit storing a plurality of description records of the game on the line; playing an individual one-feature-value extraction unit on the evening, coupling the input/polling; Save order 16 201120672 -.....1003 32826twf.doc/d 兀'Get the feature value extraction unit from the input/output interface. After the storage of a single point, the player points accumulated before the player point for the specific player Recording as a multi-segment & divide between the player to trace at least and for each of the sub-records, extracting - predicting a single 7t, the input/output embedding the respective extracted from the sub-records Two: π energy, and the wealth of the f/tt line of bird play can be output through the wheel in / out interface. The prediction system is the second game H game player behavior life phase game _, and the specific player's pre-i;.r: Please Γ 围 1 surrounding the line field player behavior described in item 10 (10) The game includes a few of the following: • Record, Role #, and “Character Performance”. The online game player behavior levy extracting unit described in the eleventh item is described in the player description record package: in addition, according to each of the sub-records, the average value of the parent-day game time of each of the sub-records τ is estimated, The game is intensive, the average time of the online time, the average of each online time, and the game time variation of at least one of the game time as the at least one feature. 13. The online game as described in claim 1 Player behavior 17 201120672 v / j\~\jy iw3 32826twf.doc/d !: Lai ^If the life related records include at least one of the following: one play poor and one player behavior 14 · as applied for patent scope The prediction system of 9 items ^, the online game player's behavior is handled from each of these sub-records. Λ At least—the eigenvalue to calculate the line of the & 疋 豕 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = Supervised learning. One of the playing games divides the player description and each of the sub-records, extracts the recorded feature value; and breaks the extracted at least-feature value to Capability. ^ (4) _ The online game's 4 - game related record, and the specific player's life home ' 18 lu03 32826twf.d 〇 c / d 201120672, : 2: ί patent scope ls items of computer program products , the game related records in 1 include at least the following:::, record, - role data H role behavior. I have 20. The computer described in claim 19, when the player tracks the track and the upper and lower pellets, the scale process ^ = each of the sub-records extracts the at least - the characteristic value when the two = The projections are calculated corresponding to each of the sub-records - the average value of each game ί, the investment game intensity - the average online time point, the = online time average, and - the daily game time variation to one As the at least one feature value. 21, as claimed in claim 18: The life-related record includes at least the following...: one material: and one player's performance. 22. The computer program product of claim 1, wherein the program command determines the female player in the future based on the at least == value extracted from each of the child records. When the possibility of leaving the game is specified, the at least one feature value extracted from each of the two records is processed by a machine learning mechanism to calculate that the specific player leaves the line within a specified period of time. The possibility of the game. 23. For computer program products as described in claim 22, the machine learning mechanism includes one of the following: a supervised learning classifier and an unsupervised learning taxonomy. 24. The computer program product of claim 23, wherein the t-learning classification includes a support vector machine. 19
TW098142520A 2009-12-11 2009-12-11 Method and system for prophesying behavior of online gamer, and computer program product thereof TW201120672A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW098142520A TW201120672A (en) 2009-12-11 2009-12-11 Method and system for prophesying behavior of online gamer, and computer program product thereof

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TW098142520A TW201120672A (en) 2009-12-11 2009-12-11 Method and system for prophesying behavior of online gamer, and computer program product thereof
US12/813,516 US20110143829A1 (en) 2009-12-11 2010-06-11 Method and system for prophesying behavior of online gamer, and computer program product thereof

Publications (1)

Publication Number Publication Date
TW201120672A true TW201120672A (en) 2011-06-16

Family

ID=44143560

Family Applications (1)

Application Number Title Priority Date Filing Date
TW098142520A TW201120672A (en) 2009-12-11 2009-12-11 Method and system for prophesying behavior of online gamer, and computer program product thereof

Country Status (2)

Country Link
US (1) US20110143829A1 (en)
TW (1) TW201120672A (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101986676B (en) * 2010-10-22 2012-10-10 浙江大学 Method for counting game time of smart mobile phone

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040039630A1 (en) * 2002-08-12 2004-02-26 Begole James M.A. Method and system for inferring and applying coordination patterns from individual work and communication activity
US7421415B2 (en) * 2004-09-07 2008-09-02 Siemens Corporate Research, Inc. Methods and systems for 3D object detection using learning
US9433868B2 (en) * 2005-12-28 2016-09-06 Palo Alto Research Center Incorporated Method, apparatus, and program product for predicting virtual presence in a persistent virtual enviornment

Also Published As

Publication number Publication date
US20110143829A1 (en) 2011-06-16

Similar Documents

Publication Publication Date Title
King et al. Unfair play? Video games as exploitative monetized services: An examination of game patents from a consumer protection perspective
Coates et al. Superstar salaries and soccer success: The impact of designated players in major league soccer
US11250347B2 (en) Personalization enhanced recommendation models
KR101766636B1 (en) Apparatus and method for player matching
US20210118085A1 (en) Methods, devices and systems for multi-player virtual hybrid wager-based and non-wager-based competitions
Barajas et al. Does sports performance influence revenues and economic results in Spanish football?
Platt et al. The role of risk preferences in pay-to-bid auctions
Gauriot et al. Fooled by performance randomness: Overrewarding luck
Gainsbury et al. Intensity and gambling harms: Exploring breadth of gambling involvement among esports bettors
JP5826967B2 (en) Method, apparatus and computer readable medium for enabling real-time competition
US20130325621A1 (en) Apparatus, system and method of providing dedicated mobile apps for use by celebrities
TW201402178A (en) Method and device for multi-level gaming competitions
Smith et al. The 9 pitfalls of data science
EP1642626B1 (en) Evaluation of possible fraud behaviour in online games
US20210398240A1 (en) System for increasing betting level of profitability and methods of use
Deutscher et al. The demand for English premier league soccer betting
Xu et al. Entrepreneurial Learning and Disincentives in Crowdfunding Markets
US10617948B2 (en) System and data collection method
TW201120672A (en) Method and system for prophesying behavior of online gamer, and computer program product thereof
Patnaik et al. A study of Prediction models for football player valuations by quantifying statistical and economic attributes for the global transfer market
Martín et al. Measuring football clubs' human capital: analytical and dynamic models based on footballers' life cycles
US20220222999A1 (en) Real time action of interest notification system
KR20150083966A (en) System and method for ranking competition game using sports match score
JP2018124677A (en) Publicly managed racing betting system
Peta Trading Bases: How a Wall Street Trader Made a Fortune Betting on Baseball