TWI776020B - Apparatus, method, and computer program product thereof for locating user interests - Google Patents
Apparatus, method, and computer program product thereof for locating user interests Download PDFInfo
- Publication number
- TWI776020B TWI776020B TW108105153A TW108105153A TWI776020B TW I776020 B TWI776020 B TW I776020B TW 108105153 A TW108105153 A TW 108105153A TW 108105153 A TW108105153 A TW 108105153A TW I776020 B TWI776020 B TW I776020B
- Authority
- TW
- Taiwan
- Prior art keywords
- user
- interest
- users
- articles
- key
- Prior art date
Links
Images
Abstract
Description
本發明係關於一種鎖定用戶興趣之裝置、方法及其電腦程式產品。具體而言,本發明係關於一種基於用戶閱讀習慣而鎖定用戶興趣之裝置、方法及其電腦程式產品。 The present invention relates to a device, method and computer program product for locking user interests. Specifically, the present invention relates to a device, method and computer program product for locking user's interests based on user's reading habits.
隨著數位時代的來臨,社會大眾已習慣在各式電子裝置上閱讀文章。許多的內容提供者會將文章分類(例如:政治類、體育類),讓使用者能方便地依據分類選擇欲閱讀的文章。另外,也有一些內容提供者會對各篇文章標註關鍵字,讓使用者能方便地檢索。 With the advent of the digital age, the general public has become accustomed to reading articles on various electronic devices. Many content providers classify articles (eg, politics, sports), so that users can easily select articles to read according to the categories. In addition, some content providers will mark each article with keywords so that users can easily retrieve it.
在此趨勢下,目前已有一些技術基於使用者的閱讀習慣提供數位服務。具體而言,某些技術係基於使用者閱讀過的文章的分類(例如:政治類、體育類)來分析使用者的興趣,但由於文章的分類過於上位,導致分析的結果會過於粗糙。另外的某些技術則是基於使用者閱讀過的文章的關鍵字來分析使用者的興趣,但各個關鍵字所傳遞的訊息太過特定,且使用者閱讀過的文章的關鍵字群又往往無法聚焦,導致無法鎖定使用者的興趣。 Under this trend, some technologies currently provide digital services based on users' reading habits. Specifically, some technologies analyze the interests of users based on the categories of articles read by the users (eg, politics, sports), but because the categories of articles are too high-level, the results of the analysis will be too rough. Some other technologies analyze the interests of users based on the keywords of the articles that the users have read, but the information conveyed by each keyword is too specific, and the keyword groups of the articles that the users have read often cannot Focus, which makes it impossible to lock the user's interest.
有鑑於此,本領域仍亟需一種能基於使用者的閱讀習慣找出 使用者的興趣的資訊探勘技術。 In view of this, there is still an urgent need in the art for a method that can find the Information mining techniques for users' interests.
為解決先前技術的間題,本發明提供一種鎖定用戶興趣之裝置、方法及其電腦程式產品。 In order to solve the problems of the prior art, the present invention provides a device, method and computer program product for locking user interests.
本發明所提供之鎖定用戶興趣之裝置包含一儲存器及一處理器,且該儲存器電性連接至該處理器。該儲存器儲存複數個興趣類別、複數篇文章、各該文章所對應之複數個關鍵字及該用戶之一閱讀記錄。該處理器利用該等文章建立複數個關聯規則(association rules),其中各該關聯規則介於該等興趣類別其中之一與該等關鍵字其中之一。該處理器還根據該閱讀記錄確認該用戶之複數篇已讀文章,且根據該用戶之該等已讀文章決定該用戶之一關鍵字集,其中該用戶之該等已讀文章為該等文章之一子集,且該用戶之該關鍵字集為該等關鍵字之一子集。該處理器還根據該用戶之該關鍵字集所對應之該等關聯規則,確認該用戶之一興趣分布。 The device for locking user interest provided by the present invention includes a storage and a processor, and the storage is electrically connected to the processor. The storage stores a plurality of interest categories, a plurality of articles, a plurality of keywords corresponding to the articles and a reading record of one of the users. The processor uses the articles to establish a plurality of association rules, wherein each association rule is between one of the interest categories and one of the keywords. The processor also confirms a plurality of read articles of the user according to the reading record, and determines a keyword set of the user according to the read articles of the user, wherein the read articles of the user are the articles a subset, and the keyword set of the user is a subset of the keywords. The processor also confirms an interest distribution of the user according to the association rules corresponding to the keyword set of the user.
本發明所提供之鎖定用戶興趣之方法適用於一電子計算裝置。該電子計算裝置儲存複數個興趣類別、複數篇文章、各該文章所對應之複數個關鍵字及一用戶之一閱讀記錄。該方法包含下列步驟:(a)利用該等文章建立複數個關聯規則,其中各該關聯規則介於該等興趣類別其中之一與該等關鍵字其中之一,(b)根據該閱讀記錄確認該用戶之複數篇已讀文章,(c)根據該用戶之該等已讀文章決定該用戶之一關鍵字集,其中該用戶之該等已讀文章為該等文章之一子集,且該用戶之該關鍵字集為該等關鍵字之一子集,以及(d)根據該用戶之該關鍵字集所對應之該等關聯規則,確認該用戶之一興趣分布。 The method for locking user interests provided by the present invention is suitable for an electronic computing device. The electronic computing device stores a plurality of interest categories, a plurality of articles, a plurality of keywords corresponding to the articles and a reading record of a user. The method includes the following steps: (a) using the articles to establish a plurality of association rules, wherein each of the association rules is between one of the interest categories and one of the keywords, (b) confirming according to the reading record a plurality of articles read by the user, (c) determine a keyword set for the user based on the articles read by the user, wherein the articles read by the user are a subset of the articles, and the article The keyword set of the user is a subset of the keywords, and (d) according to the association rules corresponding to the keyword set of the user, an interest distribution of the user is confirmed.
本發明所提供之電腦程式產品包含複數個程式指令。一電子計算裝置載入該電腦程式產品後,該電子計算裝置執行該電腦程式產品所包含之該等程式指令,因而實現一種鎖定用戶興趣之方法。該電子計算裝置儲存複數個興趣類別、複數篇文章、各該文章所對應之複數個關鍵字及一用戶之一閱讀記錄。該方法包含下列步驟:(a)利用該等文章建立複數個關聯規則,其中各該關聯規則介於該等興趣類別其中之一與該等關鍵字其中之一,(b)根據該閱讀記錄確認該用戶之複數篇已讀文章,(c)根據該用戶之該等已讀文章決定該用戶之一關鍵字集,其中該用戶之該等已讀文章為該等文章之一子集,且該用戶之該關鍵字集為該等關鍵字之一子集,以及(d)根據該用戶之該關鍵字集所對應之該等關聯規則,確認該用戶之一興趣分布。 The computer program product provided by the present invention includes a plurality of program instructions. After an electronic computing device loads the computer program product, the electronic computing device executes the program instructions contained in the computer program product, thereby realizing a method of locking user interests. The electronic computing device stores a plurality of interest categories, a plurality of articles, a plurality of keywords corresponding to the articles and a reading record of a user. The method includes the following steps: (a) using the articles to establish a plurality of association rules, wherein each of the association rules is between one of the interest categories and one of the keywords, (b) confirming according to the reading record a plurality of articles read by the user, (c) determine a keyword set for the user based on the articles read by the user, wherein the articles read by the user are a subset of the articles, and the article The keyword set of the user is a subset of the keywords, and (d) according to the association rules corresponding to the keyword set of the user, an interest distribution of the user is confirmed.
本發明所提供之興趣鎖定技術(至少包含裝置、方法及其電腦程式產品)在複數篇文章所彙整出之相異關鍵字與複數個興趣類別之間建立複數個關聯規則。在建立該等關聯規則後,本發明所提供之興趣鎖定技術便可針對單一用戶、多個用戶、一目標用戶群或/及全體用戶進行分析,以鎖定不同用戶的興趣分布,鎖定不同用戶的關鍵興趣類別,甚至針對不同興趣類別鎖定關鍵用戶。由於本發明所提供之興趣鎖定技術並非單純地根據文章的分類(例如:政治類、體育類)或單純地根據文章的關鍵字來判斷用戶之興趣分布,而是交叉比對出文章本身所具有的資訊與興趣類別間的關聯性,因此本發明所提供之興趣鎖定技術能更為準確地確認用戶之興趣分布,找出其關鍵興趣類別,且能針對不同興趣類別鎖定關鍵用戶,進而提供更為準確的數位服務。 The interest locking technology (at least including the device, the method and the computer program product thereof) provided by the present invention establishes a plurality of association rules between different keywords collected from a plurality of articles and a plurality of interest categories. After these association rules are established, the interest locking technology provided by the present invention can analyze a single user, multiple users, a target user group or/and all users, so as to lock the interest distribution of different users and lock the interests of different users. Key interest categories and even target key users for different interest categories. Because the interest locking technology provided by the present invention does not simply judge the distribution of the user's interest according to the classification of the article (for example: politics, sports) or simply according to the keywords of the article, but cross-comparisons the article itself with Therefore, the interest locking technology provided by the present invention can more accurately confirm the user's interest distribution, find out its key interest categories, and can lock key users according to different interest categories, thereby providing more accurate information. Serve for accurate digits.
以下結合圖式闡述本發明之詳細技術及實施方式,俾使本發明所屬技術領域中具有通常知識者能理解所請求保護之發明之技術特徵。 The detailed techniques and embodiments of the present invention are described below with reference to the drawings, so that those with ordinary knowledge in the technical field to which the present invention pertains can understand the technical features of the claimed invention.
1‧‧‧興趣鎖定裝置 1‧‧‧Interest Locking Device
11‧‧‧儲存器 11‧‧‧Storage
13‧‧‧處理器 13‧‧‧Processor
C1、C2、……、Cm‧‧‧興趣類別 C1, C2, ..., Cm‧‧‧Interest category
A1、A2、……、An‧‧‧文章 A1, A2, ..., An‧‧‧Articles
SA‧‧‧文章資料庫 SA‧‧‧Article Database
RR1、RR2、……、RRs‧‧‧閱讀記錄 RR1, RR2, ..., RRs‧‧‧Reading records
K1、K2、……、Kt‧‧‧關鍵字 K1, K2, ..., Kt‧‧‧Keywords
R1、R2、……、Rz‧‧‧關聯規則 R1, R2,...,Rz‧‧‧association rules
U1、U2、……、Us‧‧‧用戶 U1, U2, ..., Us‧‧‧Users
SA1、SA2、……、SAs‧‧‧複數篇已讀文章 SA1, SA2, ..., SAs‧‧‧Multiple read articles
SK1、SK2、……、SKs‧‧‧關鍵字集 SK1, SK2, ..., SKs‧‧‧Keyword Set
ID1、ID2、……、IDs‧‧‧興趣分布 ID1, ID2, ..., IDs‧‧‧Interest distribution
S201~S207‧‧‧步驟 Steps S201~S207‧‧‧
第1A圖係描繪本發明第一實施方式之興趣鎖定裝置1之架構示意圖;第1B圖係描繪關鍵字K1、K2、……、Kt及興趣類別C1、C2、……、Cm之間建立複數個關聯規則R1、R2、……、Rz之示意圖;第1C圖係描繪興趣鎖定裝置1在獲得用戶U1之興趣分布時之訊息產生流程;以及第2圖係描繪本發明第二實施方式之興趣鎖定方法之流程圖。
Fig. 1A is a schematic diagram of the structure of the
以下將透過實施方式來解釋本發明所提供之鎖定用戶興趣之裝置、方法及其電腦程式產品。然而,該等實施方式並非用以限制本發明需在如該等實施方式所述之任何環境、應用或方式方能實施。因此,關於以下實施方式之說明僅在於闡釋本發明之目的,而非用以限制本發明之範圍。應理解,在以下實施方式及圖式中,與本發明非直接相關之元件已省略而未繪示,且圖式中各元件之尺寸以及元件間之尺寸比例僅為便於繪示及說明,而非用以限制本發明之範圍。 The following will explain the device, method and computer program product of locking user interest provided by the present invention through implementation. However, these embodiments are not intended to limit the implementation of the present invention in any environment, application or manner as described in these embodiments. Therefore, the description of the following embodiments is only for the purpose of explaining the present invention, rather than limiting the scope of the present invention. It should be understood that, in the following embodiments and drawings, elements not directly related to the present invention have been omitted and not shown, and the size of each element and the size ratio between the elements in the drawings are only for convenience of illustration and description, and It is not intended to limit the scope of the present invention.
本發明之第一實施方式為一鎖定用戶興趣之裝置(下稱「興趣鎖定裝置」)1,其架構示意圖係描繪於第1A圖。興趣鎖定裝置1包含一儲存器11及一處理器13,且二者彼此電性連接。儲存器11可為一記憶體、一
通用串列匯流排(Universal Serial Bus;USB)碟、一硬碟、一光碟(Compact Disk;CD)、一隨身碟或本發明所屬技術領域中具有通常知識者所知悉之其他能儲存數位資料之非暫態儲存媒體或儲存電路。處理器13可為各種處理器、中央處理單元(Central Processing Unit;CPU)、微處理器(Microprocessor Unit;MPU)、數位訊號處理器(Digital Signal Processor;DSP)或本發明所屬技術領域中具有通常知識者所知悉之其他計算裝置。
The first embodiment of the present invention is a device for locking a user's interests (hereinafter referred to as "interest locking device") 1 , the schematic diagram of which is depicted in FIG. 1A . The
儲存器11儲存複數個興趣類別C1、C2、……、Cm。需說明者,本發明未限制興趣類別C1、C2、……、Cm之產生方式及其具體數目,其可由興趣鎖定裝置1之管理者自行建立,亦可為任何現有已清楚定義之興趣類別(例如:美商臉書公司(Facebook,Inc.)所定義之興趣類別、美商谷歌有限公司(Google LLC)所定義之興趣類別)。此外,本發明未限制興趣類別C1、C2、……、Cm儲存於儲存器11之形式。舉例而言,興趣類別C1、C2、……、Cm可被記錄於一檔案或一資料庫中,但不以此為限。
The
儲存器11還儲存複數篇文章A1、A2、……、An。文章A1、A2、……、An整體可視為一文章資料庫SA。此外,儲存器11儲存章A1、A2、……、An各自所對應之複數個關鍵字(未繪示)。於本發明的不同實施方式中,文章A1、A2、……、An各自所對應之複數個關鍵字可透過不同方式產生。
The
儲存器11還儲存複數個用戶U1、U2、……、Us所分別對應之複數個閱讀記錄RR1、RR2、……、RRs。具體而言,閱讀記錄RR1、RR2、……、RRs各自記錄所對應之用戶讀過文章A1、A2、……、An中的哪幾篇。以閱讀記錄RR1為例,其係記錄用戶U1讀過文章A1、A2、……、
An中的哪幾篇。
The
於某些實施方式中,文章A1、A2、……、An各自所對應之該等關鍵字可由一使用者(例如:文章編輯、文章作者、興趣鎖定裝置1之管理者)所直接給定的。於某些實施方式中,文章A1、A2、……、An各自所對應之該等關鍵字則可由興趣鎖定裝置1產生。舉例而言,處理器13可針對文章A1、A2、……、An個別地進行斷詞處理及停用詞(stop words)過濾,藉此得到文章A1、A2、……、An各自對應之該等關鍵字。再舉例而言,處理器13可針對文章A1、A2、……、An個別地進行斷詞處理、停用詞過濾及一詞頻-逆文件頻率(Tenn Frequency-Inverse Document Frequency;TF-IDF)演算法過濾,藉此得到文章A1、A2、……、An各自對應之該等關鍵字。本發明所屬技術領域中具有通常知識者應熟知斷詞處理、停用詞過濾與詞頻-逆文件頻率演算法之運作細節,茲不贅言。
In some embodiments, the keywords corresponding to each of the articles A1, A2, . . In some embodiments, the keywords corresponding to each of the articles A1 , A2 , . . . , An can be generated by the
於本實施方式中,處理器13利用文章A1、A2、……、An,在興趣類別C1、……、Cm與從文章A1、A2、……、An所彙整出來的相異的關鍵字之間建立複數個關聯規則(association rules)。處理器13所建立的各該關聯規則係介於興趣類別C1、C2、……、Cm其中之一與從文章A1、A2、……、An彙整得到的該等關鍵字其中之一。
In this embodiment, the
於某些實施方式中,處理器13係根據各該關鍵字與各該興趣類別C1、C2、……、Cm於文章A1、A2、……、An中同時出現之複數個比例建立該等關聯規則。若某一興趣類別與某一關鍵字同時出現之文章數目(或文章數目所佔之比例)高於一門檻值,處理器13便會在該興趣類別與該關鍵字之間建立一關聯規則。於某些實施方式中,處理器13則可根據各
該關鍵字與各該興趣類別C1、C2、……、Cm於文章A1、A2、……、An中同一段落同時出現之複數個比例建立該等關聯規則。若某一興趣類別與某一關鍵字同時出現之段落數目(或段落數目所佔之比例)高於一門檻值,處理器13便會在該興趣類別與該關鍵字之間建立一關聯規則。
In some embodiments, the
為便於理解,請參第1B圖所示之一具體範例,但其非用以限制本發明之範圍。於該具體範例中,處理器13從文章A1、A2、……、An各自對應之該等關鍵字中彙整出相異的關鍵字K1、K2、……、Kt,再利用文章A1、A2、……、An所具有之資訊於相異的關鍵字K1、K2、……、Kt與興趣類別C1、C2、……、Cm之間建立複數個關聯規則R1、R2、……、Rz。如第1B圖所示,每一條介於關鍵字與興趣類別間之直線代表一關聯規則。關聯規則R1、R2、……、Rz各自介於興趣類別C1、C2、……、Cm其中之一與關鍵字K1、K2、……、Kt其中之一。
For ease of understanding, please refer to a specific example shown in FIG. 1B, but it is not intended to limit the scope of the present invention. In this specific example, the
在建立關聯規則R1、R2、……、Rz後,處理器13便可針對單一用戶、多個用戶、一目標用戶群或/及全體用戶進行分析,以鎖定不同用戶的興趣分布,甚至鎖定不同用戶的關鍵興趣類別。
After establishing the association rules R1, R2, ..., Rz, the
茲假設欲獲得全體用戶中之用戶U1之興趣分布,興趣鎖定裝置1運作時之訊息產生流程如第1C圖所示。具體而言,處理器13會根據用戶U1之閱讀記錄RR1確認用戶U1之複數篇已讀文章SA1。用戶U1之該等已讀文章SA1為前述文章A1、A2、……、An之一子集(亦即,文章A1、A2、……、An中被用戶U1讀過的)。處理器13再根據用戶U1之該等已讀文章SA1決定用戶U1之一關鍵字集SK1。用戶U1之關鍵字集SK1為關鍵字K1、K2、……、Kt之一子集(亦即,從用戶U1之該等已讀文章SA1彙整出來的相異關鍵字
所形成之集合)。之後,處理器13根據用戶U1之關鍵字集SK1所對應之該等關聯規則(亦即,關鍵字集SK1所包含之該等關鍵字所對應之該等關聯規則,為關聯規則R1、R2、……、Rz之一子集),確認用戶U1之一興趣分布ID1。
It is assumed that the interest distribution of the user U1 among all users is to be obtained, and the message generation process when the
需說明者,興趣鎖定裝置1可採用數種不同方式呈現興趣分布ID1。於某些實施方式中,處理器13可計算用戶U1之關鍵字集SK1所包含之該等關鍵字中,屬於各興趣類別C1、C2、……、Cm之關鍵字佔關鍵字集SK1之比例,而這些比例便可視為用戶U1之一興趣分布ID1。為使興趣分布ID1更為分布,處理器13可進一步地將之正規化(normalized),使這些比例的總和為100%。再舉例而言,處理器13亦可計算用戶U1之關鍵字集SK1所包含之該等關鍵字中,屬於各興趣類別C1、C2、……、Cm之關鍵字之數目,而這些數目便可視為用戶U1之一興趣分布ID1。
It should be noted that the
若有需要,對於全體用戶中之其他用戶U2、……、Us,處理器13也可採取同樣的技術找出用戶U2、……、Us各自之興趣分布。簡言之,處理器13可根據用戶U2、……、用戶Us分別對應之閱讀記錄RR2、……、RRs確認用戶U2、……、用戶Us分別對應至複數篇已讀文章SA2、……、複數篇已讀文章SAs,其中各用戶之已讀文章為前述文章A1、A2、……、An之一子集。處理器13根據用戶U2、……、用戶Us分別對應之複數篇已讀文章SA2、……、複數篇已讀文章SAs,決定用戶U2、……、用戶Us分別對應之關鍵字集SK2、……、SKs,其中關鍵字集SK2、……、SKs各為關鍵字K1、K2、……、Kt之一子集。處理器13再根據用戶U2、……、用戶Us分別的關鍵字集SK2、……、SKs所對應之該等關聯規則,確認用戶U2、……、
用戶Us分別的興趣分布ID2、……、IDs。
If necessary, for other users U2, . In short, the
於某些實施方式中,處理器13可針對全體用戶(亦即,用戶U1、U2、……、Us)進行分析,找出全體用戶之興趣分布。具體而言,處理器13根據全體用戶之已讀文章決定該全體用戶之一關鍵字集。同理,全體用戶之該等已讀文章為前述文章A1、A2、……、An之一子集(亦即,文章A1、A2、……、An中被任一用戶讀過的),且該全體用戶之該關鍵字集為關鍵字K1、K2、……、Kt之一子集(亦即,從該全體用戶之該等已讀文章彙整出來的相異關鍵字所形成之集合)。之後,處理器13根據該全體用戶之該關鍵字集所對應之該等關聯規則,確認該全體用戶之一興趣分布(以前述任一種確認興趣分布之技術)。由該全體用戶之興趣分布可看出大眾的興趣分布。
In some embodiments, the
於某些實施方式中,處理器13可藉由比較某一用戶(例如:用戶U1)之該興趣分布及該全體用戶之該興趣分布,確認該用戶之至少一關鍵興趣類別。各該至少一關鍵興趣類別為興趣類別C1、C2、……、Cm其中之一。具體而言,若有某一(或某些)興趣類別在該用戶之興趣分布中所佔之比例高於在該全體用戶之興趣分布中所佔之比例,則那一(或那些)興趣類別為該用戶之關鍵興趣類別。舉例而言,興趣類別「閱讀」在用戶U1的興趣分布ID1中佔86.06%,而在全體用戶的興趣分布中佔32.9%,因此興趣類別「閱讀」便是為用戶U1之關鍵興趣類別。
In some embodiments, the
若有需要,對於全體用戶中之其他用戶U2、……、Us,處理器13也可採取同樣的技術,找出用戶U2、……、Us各自之至少一關鍵興趣類別。 If necessary, for other users U2, .
於某些實施方式中,處理器13可針對一目標用戶群進行分析,找出目標用戶群之興趣分布。該目標用戶群為全體用戶(亦即,用戶U1、U2、……、Us)之一子集。舉例而言,該目標用戶群可為全體用戶中曾經捐款的用戶、曾經參與某一活動的用戶,但不以此為限。
In some embodiments, the
於該等實施方式中,處理器13根據該目標用戶群之複數篇已讀文章決定該目標用戶群之一關鍵字集。該目標用戶群之該等已讀文章為前述文章A1、A2、……、An之一子集(亦即,文章A1、A2、……、An中被該目標用戶群中之任一用戶讀過的),且該目標用戶群之該關鍵字集為關鍵字K1、K2、……、Kt之一子集(亦即,從該目標用戶群之該等已讀文章彙整出來的相異關鍵字所形成之集合)。之後,處理器13根據該目標用戶群之該關鍵字集所對應之該等關聯規則,確認該目標用戶群之一興趣分布(以前述任一種確認興趣分布之技術)。
In these embodiments, the
於某些實施方式中,處理器13還可藉由比較該目標用戶群之該興趣分布及該全體用戶之該興趣分布,確認該目標用戶群之至少一關鍵興趣類別,茲不贅言。
In some embodiments, the
於某些實施方式中,處理器13還可針對該目標用戶群所包含之複數個目標用戶進行分析,找出其共通性。如前所述,該目標用戶群為全體用戶(亦即,用戶U1、U2、……、Us)之一子集,因此各該目標用戶為用戶U1、U2、……、Us其中之一。處理器13會採用前述方式確認各該目標用戶之關鍵興趣類別,再找出該等目標用戶所分別對應之該等關鍵興趣類別之一交集作為該等目標用戶之一共同關鍵興趣類別。
In some embodiments, the
舉例而言,目標用戶群包含用戶U1、用戶U2及用戶Us,其
中用戶U1的關鍵興趣類別為「閱讀」、「電影」及「遊戲」,用戶U2的關鍵興趣類別為「電影」及「運動」,而用戶Us的關鍵興趣類別為「電影」、「美食」及「旅遊」。處理器13便可找出用戶U1之關鍵興趣類別、用戶U2之關鍵興趣類別及用戶Us之關鍵興趣類別之一交集作為用戶U1、用戶U2及用戶Us的共同關鍵興趣類別(亦即,興趣類別「電影」)。
For example, the target user group includes user U1, user U2 and user Us, which
The key interest categories of user U1 are "reading", "movies" and "games", the key interest categories of user U2 are "movies" and "sports", and the key interest categories of user Us are "movies", "food" and "tourism". The
藉由前述技術,當興趣鎖定裝置1根據某一行為或某一資訊(例如:曾經捐款的用戶、曾經參與某一活動)鎖定一目標用戶群時,便能找出該目標用戶群之目標用戶之共通性。
With the aforementioned technology, when the
於某些實施方式中,處理器13還可針對興趣類別C1、C2、……、Cm中每一個,找出其所對應之至少一關鍵用戶。具體而言,處理器13先針對用戶U1、U2、……、Us中的每一個找出對應之至少一關鍵興趣類別。由於各該關鍵興趣類別為興趣類別C1、C2、……、Cm其中之一,處理器13便能根據用戶U1、U2、……、Us之該等關鍵興趣類別,確認興趣類別C1、C2、……、Cm中每一個所對應之至少一關鍵用戶。具體而言,若一興趣類別屬於某一(或某些)用戶之關鍵興趣類別,則該(等)用戶為該興趣類別之關鍵用戶。舉例而言,若興趣類別C2為用戶U1、U2、Us之關鍵興趣類別,則用戶U1、U2、Us為興趣類別C2之關鍵用戶。
In some embodiments, the
由上述說明可知,興趣鎖定裝置1在文章A1、A2、……、An之相異關鍵字與興趣類別C1、C2、……、Cm之間建立關聯規則R1、R2、……、Rz。在建立關聯規則R1、R2、……、Rz後,處理器13便可針對單一用戶、多個用戶、一目標用戶群或/及全體用戶進行分析,以鎖定不同用戶的興趣分布,鎖定不同用戶的關鍵興趣類別,甚至針對不同興趣類別
鎖定關鍵用戶。由於興趣鎖定裝置1並非單純地根據文章的分類(例如:政治類、體育類)或單純地根據文章的關鍵字來判斷用戶之興趣分布,而是交叉比對出文章本身所具有的資訊與興趣類別間的關聯性,因此興趣鎖定裝置1能更為準確地確認用戶之興趣分布,找出其關鍵興趣類別,且能針對不同興趣類別鎖定關鍵用戶。
As can be seen from the above description, the
本發明之第二實施方式為一種鎖定用戶興趣之方法(下稱「興趣鎖定方法」),其流程圖係描繪於第2圖。興趣鎖定方法適用於一電子計算裝置(例如:第一實施方式中所述之興趣鎖定裝置1),且該電子計算裝置儲存複數個興趣類別、複數篇文章、各該文章所對應之複數個關鍵字以及複數個用戶分別對應之複數個閱讀記錄。各該閱讀記錄中記載了所對應之用戶讀過該等文章中的哪幾篇。需說明者,本發明未限制該等興趣類別之產生方式及其具體數目,其可由一管理者自行建立,亦可為任何現有已清楚定義之興趣類別。
The second embodiment of the present invention is a method for locking a user's interest (hereinafter referred to as "interest locking method"), the flowchart of which is depicted in FIG. 2 . The interest locking method is suitable for an electronic computing device (for example, the
另需說明者,於某些實施方式中,各該文章所對應之該等關鍵字可由一使用者(例如:文章編輯、文章作者、管理者)所直接給定的。於某些實施方式中,各該文章所對應之該等關鍵字可由興趣鎖定方法產生。舉例而言,興趣鎖定方法可包含至少一步驟(未繪示),由該電子計算裝置針對各該文章進行斷詞處理及停用詞過濾,藉此得到各該文章所對應之該等關鍵字。再舉例而言,興趣鎖定方法可包含至少一步驟(未繪示),由該電子計算裝置針對各該文章進行斷詞處理、停用詞過濾及一詞頻-逆文件頻率演算法過濾,藉此得到各該文章所對應之該等關鍵字。 It should be noted that, in some embodiments, the keywords corresponding to each article can be directly given by a user (eg, article editor, article author, manager). In some embodiments, the keywords corresponding to each article can be generated by an interest targeting method. For example, the interest locking method may include at least one step (not shown), wherein the electronic computing device performs word segmentation processing and stop word filtering on each article, thereby obtaining the keywords corresponding to each article . For another example, the interest locking method may include at least one step (not shown) of performing word segmentation processing, stop word filtering, and word frequency-inverse document frequency algorithm filtering on each article by the electronic computing device, thereby Obtain the keywords corresponding to each article.
該興趣鎖定方法至少包含步驟S201~步驟S207。於步驟 S201,由該電子計算裝置利用該等文章建立複數個關聯規則,其中各該關聯規則介於該等興趣類別其中之一與該等關鍵字其中之一。於某些實施方式中,步驟S201係根據各該關鍵字與各該興趣類別於該等文章中同時出現之複數個比例建立該等關聯規則。於某些實施方式中,步驟S201係根據各該關鍵字與各該興趣類別於該等文章中同一段落同時出現之複數個比例建立該等關聯規則。 The interest locking method includes at least steps S201 to S207. in step S201, the electronic computing device uses the articles to establish a plurality of association rules, wherein each of the association rules is between one of the interest categories and one of the keywords. In some embodiments, step S201 establishes the association rules according to a plurality of ratios of the keywords and the interest categories appearing in the articles at the same time. In some embodiments, step S201 is to establish the association rules according to a plurality of ratios of the keywords and the interest categories appearing simultaneously in the same paragraph in the articles.
於步驟S203,由該電子計算裝置根據一用戶之一閱讀記錄確認該用戶之複數篇已讀文章,其中該用戶之該等已讀文章為該等文章之一子集。於步驟S205,由該電子計算裝置根據該用戶之該等已讀文章決定該用戶之一關鍵字集,其中該用戶之該關鍵字集為該等關鍵字之一子集。於步驟S207,由該電子計算裝置根據該用戶之該關鍵字集所對應之該等關聯規則,確認該用戶之一興趣分布。於某些實施方式中,興趣鎖定方法可針對全體用戶中的各個用戶執行步驟S203至步驟S207,藉此確認各個用戶之興趣分布。 In step S203, the electronic computing device confirms a plurality of read articles of a user according to a reading record of the user, wherein the read articles of the user are a subset of the articles. In step S205, the electronic computing device determines a keyword set of the user according to the read articles of the user, wherein the keyword set of the user is a subset of the keywords. In step S207, the electronic computing device confirms an interest distribution of the user according to the association rules corresponding to the keyword set of the user. In some embodiments, the interest locking method may perform steps S203 to S207 for each user among all users, thereby confirming the interest distribution of each user.
於某些實施方式中,該興趣鎖定方法還包含一步驟,由該電子計算裝置根據全體用戶之複數篇已讀文章決定該全體用戶之一關鍵字集。該興趣鎖定方法還包含一步驟,由該電子計算裝置根據該全體用戶之該關鍵字集所對應之該等關聯規則,確認該全體用戶之一興趣分布。 In some embodiments, the interest locking method further includes a step of determining, by the electronic computing device, a keyword set of all users according to a plurality of read articles of all users. The interest locking method further includes a step of confirming, by the electronic computing device, an interest distribution of all users according to the association rules corresponding to the keyword set of all users.
於某些實施方式中,該興趣鎖定方法還包含一步驟,由該電子計算裝置藉由比較某一用戶之該興趣分布及該全體用戶之該興趣分布,確認該用戶之至少一關鍵興趣類別。同理,若有需要,該興趣鎖定方法可針對全體用戶中的各個用戶執行此一步驟,藉此確認各該用戶之至少一關 鍵興趣分布。 In some embodiments, the interest locking method further includes a step of confirming at least one key interest category of a user by the electronic computing device by comparing the interest distribution of a user with the interest distribution of all users. Similarly, if necessary, the interest locking method can perform this step for each user among all users, thereby confirming at least one level of each user. Key interest distribution.
於某些實施方式中,該興趣鎖定方法還包含一步驟,由該電子計算裝置根據一目標用戶群之複數篇已讀文章決定該目標用戶群之一關鍵字集。該興趣鎖定方法還包含一步驟,由該電子計算裝置根據該目標用戶群之該關鍵字集所對應之該等關聯規則,確認該目標用戶群之一興趣分布,其中該目標用戶群為該全體用戶之一子集。 In some embodiments, the interest locking method further includes a step of determining, by the electronic computing device, a keyword set of the target user group according to a plurality of read articles of the target user group. The interest locking method further includes a step of confirming, by the electronic computing device, an interest distribution of the target user group according to the association rules corresponding to the keyword set of the target user group, wherein the target user group is the entire group A subset of users.
於某些實施方式中,該興趣鎖定方法還包含一步驟,由該電子計算裝置藉由比較該目標用戶群之該興趣分布及該全體用戶之該興趣分布,確認該目標用戶群之至少一關鍵興趣類別。 In some embodiments, the interest locking method further includes a step of confirming at least one key of the target user group by the electronic computing device by comparing the interest distribution of the target user group with the interest distribution of all users Interest categories.
於某些實施方式中,該興趣鎖定方法還可針對該目標用戶群所包含之複數個目標用戶進行分析,找出其共通性。該目標用戶群為全體用戶之一子集,因此各該目標用戶為前述該等用戶其中之一。該興趣鎖定方法會採用前述方式確認各該目標用戶之關鍵興趣類別,再執行一步驟以由該電子計算裝置找出該等目標用戶所分別對應之該等關鍵興趣類別之一交集作為該等目標用戶之一共同關鍵興趣類別。 In some embodiments, the interest locking method may further analyze a plurality of target users included in the target user group to find out their commonalities. The target user group is a subset of all users, so each target user is one of the aforementioned users. The interest locking method confirms the key interest categories of the target users in the aforementioned manner, and then executes a step to find, by the electronic computing device, an intersection of the key interest categories corresponding to the target users as the targets A common key interest category for one of the users.
於某些實施方式中,該興趣鎖定方法還可針對各該興趣類別,找出其所對應之至少一關鍵用戶。具體而言,該興趣鎖定方法先採用前述技術針對各該用戶找出對應之至少一關鍵興趣類別。由於各該關鍵興趣類別為該等興趣類別其中之一,該興趣鎖定方法還包含一步驟,根據該等用戶之該等關鍵興趣類別,確認各該興趣類別之至少一關鍵用戶。具體而言,若一興趣類別屬於某一(或某些)用戶之關鍵興趣類別,則該(等)用戶為該興趣類別之關鍵用戶。 In some embodiments, the interest locking method may further find at least one key user corresponding to each interest category. Specifically, the interest locking method first uses the aforementioned technology to find at least one corresponding key interest category for each user. Since each of the key interest categories is one of the interest categories, the interest locking method further includes a step of confirming at least one key user of each of the interest categories according to the key interest categories of the users. Specifically, if an interest category belongs to a certain (or some) user's key interest categories, the user(s) are the key users of the interest category.
除了上述步驟,第二實施方式能執行第一實施方式所描述之興趣鎖定裝置1之所有運作及步驟,具有同樣之功能,且達到同樣之技術效果。本發明所屬技術領域中具有通常知識者可直接瞭解第二實施方式如何基於上述第一實施方式以執行此等運作及步驟,具有同樣之功能,並達到同樣之技術效果,故不贅述。
Except for the above steps, the second embodiment can perform all the operations and steps of the
第二實施方式中所闡述之興趣鎖定方法可由包含複數個程式指令之一電腦程式產品實現。該電腦程式產品可為能被於網路上傳輸之檔案,亦可被儲存於一非暫態電腦可讀取儲存媒體中。該非暫態電腦可讀取儲存媒體可為一電子產品,例如:一唯讀記憶體(Read Only Memory;ROM)、一快閃記憶體、一軟碟、一硬碟、一光碟(Compact Disk;CD)、一數位多功能光碟(Digital Versatile Disc;DVD)、一隨身碟、一可由網路存取之資料庫或本發明所屬技術領域中具有通常知識者所知且具有相同功能之任何其他儲存媒體。該電腦程式產品所包含之該等程式指令被載入一電子計算裝置(例如:興趣鎖定裝置1)後,該電腦程式執行如在第二實施方式中所述之興趣鎖定方法。 The interest locking method described in the second embodiment can be implemented by a computer program product comprising a plurality of program instructions. The computer program product can be a file that can be transmitted over a network, or can be stored in a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium can be an electronic product, such as: a read only memory (ROM), a flash memory, a floppy disk, a hard disk, a compact disk (Compact Disk); CD), a Digital Versatile Disc (DVD), a pen drive, a network accessible database, or any other storage having the same function known to those of ordinary skill in the art to which this invention pertains media. After the program instructions included in the computer program product are loaded into an electronic computing device (eg, the interest locking device 1 ), the computer program executes the interest locking method described in the second embodiment.
需說明者,於本發明專利說明書及申請專利範圍中,某些用語(包含:用戶、子集等)前被冠以「第一」或「第二」,該等「第一」及「第二」及「第三」僅用來區分該等用語係指不同項目。 It should be noted that in the patent specification of the present invention and the scope of the patent application, certain terms (including: user, subset, etc.) are prefixed with "first" or "second", such "first" and "first" "Second" and "Third" are used only to distinguish that these terms refer to different items.
綜上所述,本發明所提供之興趣鎖定技術(至少包含裝置、方法及其電腦程式產品)在複數篇文章所彙整出之相異關鍵字與複數個興趣類別之間建立複數個關聯規則。在建立該等關聯規則後,本發明所提供之興趣鎖定技術便可針對單一用戶、多個用戶、一目標用戶群或/及全體用 戶進行分析,以鎖定不同用戶的興趣分布,鎖定不同用戶的關鍵興趣類別,甚至針對不同興趣類別鎖定關鍵用戶。由於本發明所提供之興趣鎖定技術並非單純地根據文章的分類(例如:政治類、體育類)或單純地根據文章的關鍵字來判斷用戶之興趣分布,而是交叉比對出文章本身所具有的資訊與興趣類別間的關聯性,因此本發明所提供之興趣鎖定技術能更為準確地確認用戶之興趣分布,找出其關鍵興趣類別,且能針對不同興趣類別鎖定關鍵用戶,進而提供更為準確的數位服務。 To sum up, the interest locking technology (at least including the device, method and computer program product thereof) provided by the present invention establishes a plurality of association rules between different keywords collected from a plurality of articles and a plurality of interest categories. After the association rules are established, the interest locking technology provided by the present invention can target a single user, multiple users, a target user group or/and all users Users can be analyzed to lock the interest distribution of different users, lock the key interest categories of different users, and even lock key users for different interest categories. Because the interest locking technology provided by the present invention does not simply judge the distribution of the user's interest according to the classification of the article (for example: politics, sports) or simply according to the keywords of the article, but cross-comparisons the article itself with Therefore, the interest locking technology provided by the present invention can more accurately confirm the user's interest distribution, find out its key interest categories, and can lock key users according to different interest categories, thereby providing more accurate information. Serve for accurate digits.
上述實施方式僅為例示性說明本發明之部分實施態樣,以及闡釋本發明之技術特徵,而非用來限制本發明之保護範疇及範圍。任何熟悉此技藝之人士可輕易完成之改變或均等性之安排均屬於本發明所主張之範圍,本發明之權利保護範圍應以申請專利範圍為準。 The above-mentioned embodiments are only used to illustrate some embodiments of the present invention and illustrate the technical characteristics of the present invention, but are not intended to limit the protection scope and scope of the present invention. Any changes or equality arrangements that can be easily accomplished by those skilled in the art fall within the claimed scope of the present invention, and the scope of protection of the present invention should be subject to the scope of the patent application.
A1、A2、……、An‧‧‧文章 A1, A2, ..., An‧‧‧Articles
SA‧‧‧文章資料庫 SA‧‧‧Article Database
U1、U2、……、Us‧‧‧用戶 U1, U2, ..., Us‧‧‧Users
SA1、SA2、…...、SAs‧‧‧複數篇已讀文章 SA1, SA2, ..., SAs‧‧‧Multiple read articles
SK1、SK2、……、SKs‧‧‧關鍵字集 SK1, SK2, ..., SKs‧‧‧Keyword Set
ID1、ID2、……、IDs‧‧‧興趣分布 ID1, ID2, ..., IDs‧‧‧Interest distribution
Claims (15)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108105153A TWI776020B (en) | 2019-02-15 | 2019-02-15 | Apparatus, method, and computer program product thereof for locating user interests |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108105153A TWI776020B (en) | 2019-02-15 | 2019-02-15 | Apparatus, method, and computer program product thereof for locating user interests |
Publications (2)
Publication Number | Publication Date |
---|---|
TW202032460A TW202032460A (en) | 2020-09-01 |
TWI776020B true TWI776020B (en) | 2022-09-01 |
Family
ID=73643650
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW108105153A TWI776020B (en) | 2019-02-15 | 2019-02-15 | Apparatus, method, and computer program product thereof for locating user interests |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI776020B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW200943107A (en) * | 2008-02-25 | 2009-10-16 | Yahoo Inc | Prioritizing media assets for publication |
TWI574218B (en) * | 2012-07-19 | 2017-03-11 | 菲絲博克公司 | Customizing content delivery from a brand page to a user in a social networking environment |
TWI584138B (en) * | 2013-12-04 | 2017-05-21 | 納寶股份有限公司 | System and method for providing knowledge sharing service based on user relationship information of social network service |
US20180233035A1 (en) * | 2017-02-10 | 2018-08-16 | Nec Europe Ltd. | Method and filter for floating car data sources |
-
2019
- 2019-02-15 TW TW108105153A patent/TWI776020B/en active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW200943107A (en) * | 2008-02-25 | 2009-10-16 | Yahoo Inc | Prioritizing media assets for publication |
TWI574218B (en) * | 2012-07-19 | 2017-03-11 | 菲絲博克公司 | Customizing content delivery from a brand page to a user in a social networking environment |
TWI584138B (en) * | 2013-12-04 | 2017-05-21 | 納寶股份有限公司 | System and method for providing knowledge sharing service based on user relationship information of social network service |
US20180233035A1 (en) * | 2017-02-10 | 2018-08-16 | Nec Europe Ltd. | Method and filter for floating car data sources |
Also Published As
Publication number | Publication date |
---|---|
TW202032460A (en) | 2020-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10685071B2 (en) | Methods, systems, and computer program products for storing graph-oriented data on a column-oriented database | |
TWI718643B (en) | Method and device for identifying abnormal groups | |
Boyack et al. | Co‐citation analysis, bibliographic coupling, and direct citation: Which citation approach represents the research front most accurately? | |
US20150120782A1 (en) | Systems and Methods for Identifying Influencers and Their Communities in a Social Data Network | |
US9602513B2 (en) | Access control of edges in graph index applications | |
US10657186B2 (en) | System and method for automatic document classification and grouping based on document topic | |
CN110362829B (en) | Quality evaluation method, device and equipment for structured medical record data | |
US20140324965A1 (en) | Recommending media items based on purchase history | |
JP7332949B2 (en) | Evaluation method, evaluation program, and information processing device | |
TW201820173A (en) | De-identification data generation apparatus, method, and computer program product thereof | |
CN108009223B (en) | Method and device for detecting consistency of transaction data | |
US10430454B2 (en) | Systems and methods for culling search results in electronic discovery | |
US11176196B2 (en) | Unified pipeline for media metadata convergence | |
US9734229B1 (en) | Systems and methods for mining data in a data warehouse | |
CN110019017B (en) | High-energy physical file storage method based on access characteristics | |
WO2019128317A1 (en) | Article pushing method, device, server, computing device and storage medium | |
TWI776020B (en) | Apparatus, method, and computer program product thereof for locating user interests | |
US20180096436A1 (en) | Computing System for Automatically Obtaining Age Data in a Social Data Network | |
CN107404491A (en) | Terminal environments method for detecting abnormality, detection means and computer-readable recording medium | |
US20180349372A1 (en) | Media item recommendations based on social relationships | |
Sun | Topic modeling and spam detection for short text segments in web forums | |
US11567906B2 (en) | Generation and traversal of a hierarchical index structure for efficient data retrieval | |
US9898485B2 (en) | Dynamic context-based data protection and distribution | |
La Morgia et al. | TGDataset: a Collection of Over One Hundred Thousand Telegram Channels | |
JP2010238041A (en) | Classification system revision support program, classification system revision support device and classification system revision support method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GD4A | Issue of patent certificate for granted invention patent |