本發明係關於一種提供推薦內容之電子裝置及方法。具體而言，關於一種基於用戶之情緒狀態來確定並提供針對用戶之推薦內容之技術。The present invention relates to an electronic device and method for providing recommended content. Specifically, it relates to a technology for determining and providing recommended content for a user based on the user's emotional state.
最近，隨著OTT(over the top，附加服務)服務大眾化，用戶可藉由所擁有之終端裝置來利用OTT服務。用戶可藉由終端裝置而不受時間與空間之制約地獲得各種內容。OTT服務提供者持續努力向用戶提供適合之推薦內容，以引導用戶長期使用OTT服務。Recently, with the popularization of OTT (over the top, additional services), users can use OTT services through their own terminal devices. Users can obtain various contents through terminal devices without being restricted by time and space. OTT service providers continue to work hard to provide users with suitable recommended contents to guide users to use OTT services for a long time.
OTT服務提供者一直以通常之方式提供推薦內容，而並非向用戶提供個性化之推薦內容。例如，一直基於用戶過去之視聽歷史來提供推薦內容。例如，一直將目前點閱次數最高之內容作為推薦內容來提供，或將由OTT服務提供者確定之內容作為推薦內容來提供。OTT service providers have been providing recommended content in a conventional manner, rather than providing personalized recommended content to users. For example, recommended content has been provided based on the user's past viewing history. For example, the content with the highest number of clicks at present has been provided as recommended content, or the content determined by the OTT service provider has been provided as recommended content.
[發明所欲解決之問題][The problem the invention is trying to solve]
根據本發明之各種實施例，其欲解決之技術問題在於：分析用戶之情緒狀態，基於用戶之情緒狀態來確定並提供針對用戶之推薦內容。According to various embodiments of the present invention, the technical problem to be solved is: analyzing the emotional state of the user, and determining and providing recommended content for the user based on the emotional state of the user.
根據本發明之各種實施例，其欲解決之技術問題在於：確定適於複數個內容各者之複數個標籤，基於複數個標籤及用戶之情緒狀態來確定並提供針對用戶之推薦內容。進而，欲解決之技術問題在於：利用人工智慧模型來獲得適於複數個內容各者之複數個標籤，藉此可對複數個內容各者賦予更準確之標籤。 [解決問題之技術手段]According to various embodiments of the present invention, the technical problem to be solved is: determining multiple labels suitable for each of multiple contents, and determining and providing recommended content for the user based on the multiple labels and the user's emotional state. Furthermore, the technical problem to be solved is: using an artificial intelligence model to obtain multiple labels suitable for each of multiple contents, thereby giving more accurate labels to each of the multiple contents.[Technical means for solving the problem]
本發明之各種實施例之電子裝置可包括：通訊電路，其與複數個用戶各者之終端裝置通訊連接；記憶體，其儲存分類為複數個類型之複數個內容；及處理器。各種實施例之上述處理器能夠以如下方式構成：確認第1用戶之內容視聽歷史；基於上述第1用戶之內容視聽歷史，確定上述複數個類型中之上述第1用戶偏好之至少一個第1類型；確認上述第1用戶是否連續視聽屬於與上述至少一個第1類型不同之第2類型之至少一個內容預定期間；響應於上述第1用戶連續視聽屬於上述第2類型之至少一個內容預定期間，基於上述第2類型來分析上述第1用戶之情緒狀態；基於上述第1用戶之情緒狀態，確定上述複數個內容中之針對上述第1用戶之推薦內容；向上述第1用戶之終端裝置傳輸確定之上述推薦內容。The electronic device of various embodiments of the present invention may include: a communication circuit, which is connected to the terminal devices of each of a plurality of users; a memory, which stores a plurality of contents classified into a plurality of types; and a processor. The processor of various embodiments can be configured as follows: confirming the content viewing history of the first user; determining at least one first type preferred by the first user among the plurality of types based on the content viewing history of the first user; confirming whether the first user has continuously viewed at least one content of the second type that is different from the at least one first type for a predetermined period of time; in response to the first user continuously viewing at least one content of the second type for a predetermined period of time, analyzing the emotional state of the first user based on the second type; determining recommended content for the first user among the plurality of contents based on the emotional state of the first user; and transmitting the determined recommended content to the terminal device of the first user.
各種實施例之上述處理器能夠以如下方式構成：自第1外部伺服器收集上述複數個內容各者之資訊；基於上述複數個內容各者之資訊，確定適於上述複數個內容各者之複數個標籤；對上述複數個內容分別賦予上述複數個標籤；基於上述複數個標籤，確定上述複數個內容各者之類型。The above-mentioned processor of various embodiments can be constructed in the following manner: collecting information of each of the above-mentioned multiple contents from the first external server; determining multiple tags suitable for each of the above-mentioned multiple contents based on the information of each of the above-mentioned multiple contents; assigning the above-mentioned multiple tags to the above-mentioned multiple contents respectively; and determining the type of each of the above-mentioned multiple contents based on the above-mentioned multiple tags.
各種實施例之上述記憶體可儲存人工智慧模型，上述處理器以如下方式構成：向上述人工智慧模型輸入上述複數個內容各者之資訊；自上述人工智慧模型獲得適於上述複數個內容各者之複數個標籤；對上述複數個內容分別賦予所獲得之上述複數個標籤；且上述人工智慧模型為以如下方式構建之相關模型：利用機器學習演算法來對輸入資料集與輸出資料集之間的相關關係進行建模，該輸入資料集包括複數個樣本內容各者之資訊，該輸出資料集包括適於上述複數個樣本內容各者之複數個標籤。The memory of various embodiments can store an artificial intelligence model, and the processor is constructed in the following manner: information of each of the plurality of contents is input into the artificial intelligence model; a plurality of labels suitable for each of the plurality of contents are obtained from the artificial intelligence model; the plurality of labels obtained are respectively assigned to the plurality of contents; and the artificial intelligence model is a correlation model constructed in the following manner: a machine learning algorithm is used to model the correlation between an input data set and an output data set, wherein the input data set includes information of each of the plurality of sample contents, and the output data set includes a plurality of labels suitable for each of the plurality of sample contents.
各種實施例之上述第1外部伺服器可為運營門戶網站之伺服器，上述處理器以如下方式構成：自上述第1外部伺服器爬取上述複數個內容各者之概要、評論及元資料，藉此收集上述複數個內容各者之資訊。The first external server of various embodiments may be a server for operating a portal website, and the processor is configured as follows: crawling the summary, comments and metadata of each of the plurality of contents from the first external server, thereby collecting information of each of the plurality of contents.
各種實施例之上述處理器能夠以如下方式構成：基於上述第1用戶之內容視聽歷史，確認上述複數個類型各者之內容視聽次數；按照內容視聽次數由多到少之順序來確定上述複數個類型中之上述至少一個第1類型。The processor of various embodiments can be configured as follows: based on the content viewing history of the first user, confirm the number of times the content is viewed for each of the plurality of types; and determine at least one first type among the plurality of types in order of the number of times the content is viewed.
各種實施例之上述第1用戶之情緒狀態可屬於預定之複數個情緒狀態中之一者，上述複數個情緒狀態各者以與至少一個類型對應之方式設定。The emotional state of the first user in various embodiments may belong to one of a plurality of predetermined emotional states, each of which is set in a manner corresponding to at least one type.
各種實施例之上述處理器能夠以如下方式構成：連同確定之上述推薦內容一併向上述第1用戶之終端裝置傳輸上述第1用戶之情緒狀態的資訊。The processor of various embodiments can be configured as follows: transmitting information on the emotional state of the first user together with the determined recommended content to the terminal device of the first user.
各種實施例之上述處理器能夠以如下方式構成：針對上述複數個內容各者提取至少一個影像；獲得上述複數個用戶對上述複數個內容各者之評價資訊；基於上述至少一個影像及上述複數個用戶之評價資訊，確定適於上述複數個內容各者之複數個標籤；對上述複數個內容分別賦予上述複數個標籤；基於上述複數個標籤，確定上述複數個內容各者之類型。The above-mentioned processor of various embodiments can be constructed in the following manner: extract at least one image for each of the above-mentioned multiple contents; obtain the evaluation information of the above-mentioned multiple users on each of the above-mentioned multiple contents; determine a plurality of labels suitable for each of the above-mentioned multiple contents based on the above-mentioned at least one image and the evaluation information of the above-mentioned multiple users; assign the above-mentioned multiple labels to the above-mentioned multiple contents respectively; and determine the type of each of the above-mentioned multiple contents based on the above-mentioned multiple labels.
各種實施例之上述通訊電路可與提供天氣資訊之第2外部伺服器通訊連接，上述處理器以如下方式構成：自上述第1用戶之終端裝置收集上述第1用戶之位置資訊；自上述第2外部伺服器收集與上述第1用戶之位置資訊對應之天氣資訊；基於上述第2類型及上述天氣資訊，分析上述第1用戶之情緒狀態。The communication circuit of various embodiments can be connected to a second external server that provides weather information, and the processor is constructed in the following manner: collecting the location information of the first user from the terminal device of the first user; collecting weather information corresponding to the location information of the first user from the second external server; and analyzing the emotional state of the first user based on the second type and the weather information.
各種實施例之上述複數個內容可為視訊內容。The aforementioned multiple contents in various embodiments may be video contents.
本發明之各種實施例之藉由電子裝置而實行之提供推薦內容的方法可包括如下動作：確認第1用戶之內容視聽歷史；基於上述第1用戶之內容視聽歷史，確定複數個類型中之上述第1用戶偏好之至少一個第1類型；確認上述第1用戶是否連續視聽屬於與上述至少一個第1類型不同之第2類型之至少一個內容預定期間；響應於上述第1用戶連續視聽屬於上述第2類型之至少一個內容預定期間，基於上述第2類型來分析上述第1用戶之情緒狀態；基於上述第1用戶之情緒狀態，確定複數個內容中之針對上述第1用戶之推薦內容；及向上述第1用戶之終端裝置傳輸上述確定之推薦內容。 [發明之效果]The method for providing recommended content implemented by an electronic device in various embodiments of the present invention may include the following actions: confirming the content viewing history of a first user; determining at least one first type preferred by the first user among a plurality of types based on the content viewing history of the first user; confirming whether the first user continuously views a second type different from the at least one first type. at least one predetermined content period; in response to the first user continuously viewing at least one predetermined content period belonging to the second type, analyzing the emotional state of the first user based on the second type; determining recommended content for the first user from a plurality of contents based on the emotional state of the first user; and transmitting the determined recommended content to the terminal device of the first user.[Effect of the invention]
根據本發明之各種實施例，可分析用戶之情緒狀態，基於用戶之情緒狀態來確定並提供針對用戶之推薦內容。According to various embodiments of the present invention, the emotional state of a user may be analyzed, and recommended content may be determined and provided to the user based on the emotional state of the user.
根據本發明之各種實施例，可確定適於複數個內容各者之複數個標籤，基於複數個標籤及用戶之情緒狀態來確定並提供針對用戶之推薦內容。進而，利用人工智慧模型來獲得適於複數個內容各者之複數個標籤，藉此可對複數個內容各者賦予更準確之標籤。According to various embodiments of the present invention, a plurality of labels suitable for each of a plurality of contents can be determined, and based on the plurality of labels and the user's emotional state, recommended content for the user can be determined and provided. Furthermore, an artificial intelligence model is used to obtain a plurality of labels suitable for each of a plurality of contents, thereby giving a more accurate label to each of the plurality of contents.
本發明之實施例係以對本發明之技術思想進行說明為目的而例示者。本發明之發明申請專利範圍並不限定於以下提出之實施例或對該等實施例之具體說明。The embodiments of the present invention are provided for the purpose of illustrating the technical concept of the present invention. The scope of the invention application of the present invention is not limited to the embodiments presented below or the specific description of the embodiments.
若無其他定義，則本發明中使用之所有技術用語及科學用語具有本發明所屬技術領域中具有常識者通常理解之含義。本發明中使用之所有用語係以更明確地說明本發明為目的而選擇，並非為了限制本發明之發明申請專利範圍而選擇。Unless otherwise defined, all technical and scientific terms used in this invention have the meanings commonly understood by those with common sense in the technical field to which this invention belongs. All terms used in this invention are selected for the purpose of more clearly describing this invention and are not selected to limit the scope of the invention application of this invention.
關於本發明中使用之如「包括」、「具備」、「具有」、等表述，除非於包括相應表述之語句或文章中另有提及，否則應理解為具有包括其他實施例之可能性之開放型用語(open-ended terms)。Expressions such as “including,” “having,” and “having” used in the present invention should be understood as open-ended terms that have the possibility of including other embodiments, unless otherwise mentioned in the sentence or text including the corresponding expression.
除非另有提及，否則本發明中記述之單數型之表述可包括複數型之含義，這同樣用於發明申請專利範圍中記載之單數型之表述。Unless otherwise mentioned, the singular expressions described in the present invention may include the plural meaning, and the same applies to the singular expressions described in the scope of the invention application.
本發明中使用之「第1」、「第2」等表述係為了相互區分複數個構成要素而使用，並非限定該等構成要素之順序或重要度。The expressions "first", "second", etc. used in the present invention are used to distinguish a plurality of components from each other, and do not limit the order or importance of the components.
本發明中使用之用語「部」係指指軟體、或如FPGA(field-programmable gate array，現場可程式閘陣列)、ASIC(application specific integrated circuit，特殊應用積體電路)之硬體構成要素。然而，「部」並不限定於軟體或硬體。「部」能夠以位於可尋址之儲存媒體之方式構成，亦能夠以再生一個或一個以上之處理器之方式構成。因此，作為一例，「部」包括如軟體構成要素、物件導向軟體構成要素、類構成要素及任務構成要素之構成要素、流程、函數、屬性、程式、次常式、程式碼之片段、驅動器、韌體、微碼、電路、資料、資料庫、資料構造、表、陣列及變量。構成要素與「部」之內部提供之功能可由更少數量之構成要素及「部」結合，亦可進而分離成另外之構成要素及「部」。The term "component" used in the present invention refers to software, or hardware components such as FPGA (field-programmable gate array) and ASIC (application specific integrated circuit). However, "component" is not limited to software or hardware. "Component" can be configured in a manner located in an addressable storage medium, and can also be configured in a manner to reproduce one or more processors. Therefore, as an example, "component" includes components such as software components, object-oriented software components, class components and task components, processes, functions, properties, programs, subroutines, code segments, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays and variables. The functions provided by the components and parts can be combined with a smaller number of components and parts, or can be further separated into additional components and parts.
本發明中使用之所謂「基於～」之表述係用於記述包含該表述之語句或文章中記述之對確定、判斷之行為或動作產生影響之一個以上的因素，該表述不排除對確定、判斷之行為或動作產生影響之另外之因素。The expression "based on..." used in the present invention is used to describe one or more factors that affect the behavior or action of determination or judgment described in the sentence or article containing the expression, and the expression does not exclude other factors that affect the behavior or action of determination or judgment.
於本發明中，在提到某個構成要素「連接」或「連結」於另一構成要素之情形時，應理解為上述某個構成要素可直接連接或連結於上述另一構成要素、或能夠以新的其他構成要素為介質而連接或連結於上述另一構成要素。In the present invention, when it is mentioned that a certain constituent element is "connected" or "linked" to another constituent element, it should be understood that the above-mentioned certain constituent element can be directly connected or linked to the above-mentioned other constituent element, or can be connected or linked to the above-mentioned other constituent element through a new other constituent element.
以下，參照附圖對本發明之實施例進行說明。於附圖中，對相同或對應之構成要素賦予相同之參照符號。又，於以下實施例之說明中，可省略對相同或對應之構成要素之重複記述。然而，即使省略有關構成要素之記述，亦並不意味此種構成要素不包括於某一實施例中。Hereinafter, the embodiments of the present invention will be described with reference to the attached drawings. In the attached drawings, the same reference symbols are given to the same or corresponding constituent elements. In addition, in the following description of the embodiments, the repeated description of the same or corresponding constituent elements may be omitted. However, even if the description of the relevant constituent elements is omitted, it does not mean that such constituent elements are not included in a certain embodiment.
圖1係表示本發明之各種實施例之運營OTT(over the top)服務之系統10的圖。系統10可包括電子裝置110及複數個終端裝置120。電子裝置110可為用以運營OTT(over the top)服務之伺服器(server)裝置。本文中使用之複數個終端裝置120之用語可指複數個用戶各者之終端裝置。複數個用戶可指為了利用OTT服務而加入之用戶。FIG. 1 is a diagram of a system 10 for operating an OTT (over the top) service according to various embodiments of the present invention. The system 10 may include an electronic device 110 and a plurality of terminal devices 120. The electronic device 110 may be a server device for operating an OTT (over the top) service. The term “a plurality of terminal devices 120” used herein may refer to a terminal device of each of a plurality of users. A plurality of users may refer to users who join in order to utilize the OTT service.
系統10可與複數個外部伺服器(未表示)通訊連接。外部伺服器例如可包括運營門戶網站之第1外部伺服器、及提供天氣資訊之第2外部伺服器。The system 10 can be connected to a plurality of external servers (not shown) for communication. The external servers may include, for example, a first external server for operating a portal website and a second external server for providing weather information.
複數個用戶各者之終端裝置120可指用以利用OTT服務之用戶(或消費者)之裝置。終端裝置120可為能夠連接於網際網路之裝置，例如，可為行動電話、智慧型手機、筆記型電腦、可穿戴裝置或HMD(head mounted display，頭戴式顯示器)。複數個用戶各者使用之複數個終端裝置120可藉由網路而與電子裝置110彼此連接來收發各種資料。The terminal device 120 of each of the plurality of users may refer to a device of a user (or consumer) for utilizing the OTT service. The terminal device 120 may be a device capable of connecting to the Internet, for example, a mobile phone, a smart phone, a laptop, a wearable device, or an HMD (head mounted display). The plurality of terminal devices 120 used by each of the plurality of users may be connected to the electronic device 110 via a network to send and receive various data.
電子裝置110可藉由與OTT服務相關之應用程式或網站，向終端裝置120提供複數個內容。複數個內容例如可為視訊內容。電子裝置110可儲存複數個內容，複數個內容可分類為複數個類型(genre)。例如，電子裝置110可儲存分類為「動作(action)」類型之第1內容、分類為「喜劇(comedy)」類型之第2內容、分類為「紀錄片(documentary)」類型之第3內容、及分類為「恐怖片(horror)」類型之第4內容。上述類型為例示性者，當然可包括更多之類型。The electronic device 110 can provide a plurality of contents to the terminal device 120 through an application or website related to the OTT service. The plurality of contents can be, for example, video contents. The electronic device 110 can store a plurality of contents, and the plurality of contents can be classified into a plurality of genres. For example, the electronic device 110 can store a first content classified as an "action" genre, a second content classified as a "comedy" genre, a third content classified as a "documentary" genre, and a fourth content classified as a "horror" genre. The above genres are exemplary and can of course include more genres.
於本圖中，為了便於說明，假設複數個終端裝置120包括第1用戶之終端裝置120a、第2用戶之終端裝置120b及第3用戶之終端裝置120c來進行說明，然而，用戶終端裝置之數量亦可為4個以上。In this figure, for the sake of convenience, it is assumed that the plurality of terminal devices 120 include a first user's terminal device 120a, a second user's terminal device 120b, and a third user's terminal device 120c. However, the number of user terminal devices may be more than four.
各種實施例之電子裝置110可收集並儲存複數個用戶之個人資訊。電子裝置110可儲存加入OTT服務之第1用戶之個人資訊。各種實施例之電子裝置110可儲存複數個用戶之內容視聽歷史。電子裝置110可儲存加入OTT服務之第1用戶之內容視聽歷史。The electronic device 110 of various embodiments can collect and store personal information of multiple users. The electronic device 110 can store personal information of the first user who joins the OTT service. The electronic device 110 of various embodiments can store the content viewing history of multiple users. The electronic device 110 can store the content viewing history of the first user who joins the OTT service.
各種實施例之電子裝置110可收集並儲存複數個用戶之與時間對應之位置資訊。電子裝置110可自加入OTT服務之第1用戶之終端裝置120a收集預定期間內之第1用戶之與時間對應的位置資訊。於收集位置資訊之前，需先藉由第1用戶之終端裝置120a而得到第1用戶對收集位置資訊之同意。The electronic device 110 of various embodiments can collect and store location information corresponding to time of multiple users. The electronic device 110 can collect the location information corresponding to time of the first user within a predetermined period from the terminal device 120a of the first user who joins the OTT service. Before collecting the location information, the first user's consent to the collection of the location information must be obtained through the terminal device 120a of the first user.
圖2係表示本發明之各種實施例之電子裝置110及終端裝置120之方塊圖。FIG. 2 is a block diagram showing an electronic device 110 and a terminal device 120 according to various embodiments of the present invention.
參照圖2，各種實施例之電子裝置110可包括處理器111、記憶體113及通訊電路115。可省略電子裝置110所包括之構成要素中之至少一者、或將其他構成要素追加至電子裝置110。可附加地或替代地整合一部分之構成要素而實現、或實現為單個或複數個個體。電子裝置110內之至少一部分構成要素可藉由匯流排(bus)、GPIO(general purpose input/output，通用目的輸入輸出)、SPI(serial peripheral interface，串列周邊介面)或MIPI(mobile industry processor interface，行動產業處理器介面)等而彼此連接，從而收發資料及/或信號。2 , the electronic device 110 of various embodiments may include a processor 111, a memory 113, and a communication circuit 115. At least one of the components included in the electronic device 110 may be omitted, or other components may be added to the electronic device 110. A portion of the components may be integrated additionally or alternatively to be implemented, or may be implemented as a single or multiple entities. At least a portion of the components in the electronic device 110 may be connected to each other via a bus, GPIO (general purpose input/output), SPI (serial peripheral interface), or MIPI (mobile industry processor interface), etc., to transmit and receive data and/or signals.
根據各種實施例，電子裝置110之處理器111可為能夠實行電子裝置110之各構成要素(例如，記憶體113)之控制及/或與通訊相關之運算或資料處理的構成。例如，處理器111可與電子裝置110之構成要素作動地連接。處理器111可將自電子裝置110之其他構成要素接收之命令或資料加載(load)至記憶體113，對儲存於記憶體113之命令或資料進行處理，並儲存結果資料。本文中揭示之處理器111亦可指一個以上之處理器111之集合。根據各種實施例，電子裝置110之記憶體113可儲存針對處理器111之動作之指令。According to various embodiments, the processor 111 of the electronic device 110 may be a component capable of controlling each component (e.g., memory 113) of the electronic device 110 and/or performing operations or data processing related to communication. For example, the processor 111 may be operatively connected to the components of the electronic device 110. The processor 111 may load commands or data received from other components of the electronic device 110 to the memory 113, process the commands or data stored in the memory 113, and store the resultant data. The processor 111 disclosed herein may also refer to a collection of more than one processor 111. According to various embodiments, the memory 113 of the electronic device 110 may store instructions for actions of the processor 111.
根據各種實施例，電子裝置110之通訊電路115可與外部裝置(例如，終端裝置120)建立有線或無線通訊通道來與外部裝置收發各種資料。根據一實施例，為了與外部裝置進行有線通訊，通訊電路115可包括用以藉由有線纜線而與外部裝置連接之至少一個埠。於上述情形時，通訊電路115可藉由至少一個埠而與有線連接之外部裝置實行通訊。根據一實施例，通訊電路115能夠以如下方式構成：包括蜂巢式通訊模組而連接於蜂巢網路(例如，3G(3rd Mobile Communication Technology，第三代行動通訊技術)、LTE(Long-Term Evolution，長期演進)、5G(5th Generation Mobile Communication Technology，第五代行動通訊技術)、Wibro(Wireless Broadband，無線寬頻)或Wimax(Worldwide Interoperability for Microwave Access，全球互通微波存取))。根據各種實施例，通訊電路115可包括近距離通訊模組而利用近距離通訊(例如，Wi-Fi(Wireless Fidelity，無線保真)、藍牙(Bluetooth)、低功耗藍牙(Bluetooth Low Energy，BLE)、UWB(Ultra Wide Band，超寬頻))來與外部裝置收發資料，但並不限制於此。根據一實施例，通訊電路115可包括用以進行非接觸式通訊之非接觸通訊模組。非接觸式通訊例如可包括如NFC(near field communication，近場通訊)通訊、RFID(radio frequency identification，無線射頻識別)通訊或MST(magnetic secure transmission，磁性安全傳輸)通訊之至少一種非接觸方式之短距離通訊技術。According to various embodiments, the communication circuit 115 of the electronic device 110 can establish a wired or wireless communication channel with an external device (e.g., the terminal device 120) to send and receive various data with the external device. According to one embodiment, in order to perform wired communication with the external device, the communication circuit 115 may include at least one port for connecting to the external device via a wired cable. In the above case, the communication circuit 115 can communicate with the wired external device via at least one port. According to one embodiment, the communication circuit 115 can be configured as follows: including a cellular communication module and connected to a cellular network (e.g., 3G (3rd Mobile Communication Technology), LTE (Long-Term Evolution), 5G (5th Generation Mobile Communication Technology), Wibro (Wireless Broadband) or Wimax (Worldwide Interoperability for Microwave Access)). According to various embodiments, the communication circuit 115 may include a short-range communication module and utilize short-range communication (e.g., Wi-Fi (Wireless Fidelity), Bluetooth, Bluetooth Low Energy (BLE), UWB (Ultra Wide Band)) to transmit and receive data with an external device, but is not limited thereto. According to one embodiment, the communication circuit 115 may include a contactless communication module for contactless communication. The contactless communication may include, for example, at least one contactless short-distance communication technology such as NFC (near field communication), RFID (radio frequency identification), or MST (magnetic secure transmission).
各種實施例之終端裝置120可包括控制器121、記憶體123、顯示器125及通訊電路127。即使省略或置換圖2所示之構成中之一部分，亦不會阻礙實現本文中揭示之各種實施例。又，本文中揭示之第1用戶之終端裝置120a與本圖中說明之終端裝置120相同。The terminal device 120 of various embodiments may include a controller 121, a memory 123, a display 125, and a communication circuit 127. Even if a part of the components shown in FIG. 2 is omitted or replaced, it will not hinder the implementation of various embodiments disclosed herein. In addition, the terminal device 120a of the first user disclosed herein is the same as the terminal device 120 described in this figure.
各種實施例之終端裝置120之控制器121可為如下之構成：實行終端裝置120之各構成要素之控制及/或與通訊相關之運算或資料處理。例如，控制器121可與終端裝置120之構成要素作動地連接。控制器121可將自終端裝置120之其他構成要素接收之命令或資料加載(load)至終端裝置120之記憶體123，對儲存於記憶體123之命令或資料進行處理，並儲存結果資料。各種實施例之記憶體123可儲存針對控制器121之動作之指令。The controller 121 of the terminal device 120 of various embodiments may be configured as follows: to implement control of each component of the terminal device 120 and/or operations or data processing related to communication. For example, the controller 121 may be operatively connected to the components of the terminal device 120. The controller 121 may load commands or data received from other components of the terminal device 120 to the memory 123 of the terminal device 120, process the commands or data stored in the memory 123, and store the result data. The memory 123 of various embodiments may store instructions for the actions of the controller 121.
各種實施例之終端裝置120之顯示器125可基於控制器121之控制而顯示各種畫面。顯示器125能夠以觸控感測器面板(touch sensor panel，TSP)之形態實現，該觸控感測器面板可識別各種外部對象(例如，手指)之接觸或靠近。觸控感測器面板可具有各種構造及類型，本發明中可應用所有構造及類型之觸控感測器面板。顯示器125可包括靜電電容式感測器以識別各種外部對象之接觸或靠近。靜電電容式感測器可包括複數個電容器，靜電電容式感測器可對電容器施加電信號。電容器可對應於電信號之施加而充電及放電。若對電容器施加電信號，則可根據電信號之電壓大小而對電容器進行充電。顯示器125可基於電容式感測器收集之信號來接收觸控輸入。例如，顯示器125可接收用戶之輕觸(tap)觸控、雙輕觸(double tap)觸控、滑動(sliding)觸控、拖放(drag&drop)觸控及長(long)觸控。The display 125 of the terminal device 120 of various embodiments can display various screens based on the control of the controller 121. The display 125 can be implemented in the form of a touch sensor panel (TSP), which can identify the contact or approach of various external objects (e.g., fingers). The touch sensor panel can have various structures and types, and the present invention can be applied to touch sensor panels of all structures and types. The display 125 may include an electrostatic capacitive sensor to identify the contact or approach of various external objects. The electrostatic capacitive sensor may include a plurality of capacitors, and the electrostatic capacitive sensor may apply an electrical signal to the capacitor. The capacitor may be charged and discharged in response to the application of the electrical signal. If an electrical signal is applied to the capacitor, the capacitor can be charged according to the voltage of the electrical signal. The display 125 can receive touch input based on the signal collected by the capacitive sensor. For example, the display 125 can receive a user's tap touch, double tap touch, sliding touch, drag & drop touch, and long touch.
圖3係本發明之各種實施例之電子裝置110之動作流程圖。FIG. 3 is a flowchart of the operation of the electronic device 110 according to various embodiments of the present invention.
參照動作流程圖300，於動作310中，各種實施例之電子裝置110之處理器111可確認第1用戶之內容視聽歷史。電子裝置110之記憶體113可儲存分類為複數個類型之複數個內容，儲存複數個用戶各者之內容視聽歷史。處理器111可基於記憶體113中儲存之資訊，確認第1用戶之內容視聽歷史。第1用戶之內容視聽歷史可指與第1用戶視聽之內容相關之記錄。複數個內容例如可為視訊內容。Referring to the action flow chart 300, in action 310, the processor 111 of the electronic device 110 of various embodiments may confirm the content viewing history of the first user. The memory 113 of the electronic device 110 may store a plurality of contents classified into a plurality of types, and store the content viewing history of each of the plurality of users. The processor 111 may confirm the content viewing history of the first user based on the information stored in the memory 113. The content viewing history of the first user may refer to records related to the content viewed by the first user. The plurality of contents may be, for example, video content.
於動作320中，各種實施例之處理器111可基於第1用戶之內容視聽歷史，確定複數個類型中之上述第1用戶偏好之至少一個第1類型。例如，在基於第1用戶之內容視聽歷史時，若第1用戶主要視聽「動作」類型及「喜劇」類型之內容，則處理器111可確定第1用戶偏好之至少一個第1類型為「動作」類型及「喜劇」類型。第1用戶偏好之第1類型之數量可為一個，亦可為2個以上。In action 320, the processor 111 of various embodiments may determine at least one first type preferred by the first user among the plurality of types based on the first user's content viewing history. For example, based on the first user's content viewing history, if the first user mainly views and listens to content of the "action" type and the "comedy" type, the processor 111 may determine that at least one first type preferred by the first user is the "action" type and the "comedy" type. The number of first types preferred by the first user may be one or more than two.
處理器111可基於第1用戶之內容視聽歷史，確認複數個類型各者之內容視聽次數。處理器111可基於複數個類型各者之內容視聽次數，按照內容視聽次數由多到少之順序來確定複數個類型中之至少一個第1類型。例如，在基於第1用戶之內容視聽歷史時，若第1用戶視聽「動作」類型之內容最多，其次為「喜劇」類型之內容，繼而為「恐怖片」類型之內容，則處理器111可確定第1用戶偏好之至少一個第1類型為「動作」類型、「喜劇」類型及「恐怖片」類型。The processor 111 may confirm the number of times the content of each of the plurality of genres has been viewed based on the content viewing history of the first user. The processor 111 may determine at least one first genre among the plurality of genres in the order of the number of times the content has been viewed from most to least based on the content viewing history of the first user. For example, based on the content viewing history of the first user, if the first user has viewed the most content of the “action” genre, followed by the “comedy” genre, and then the “horror movie” genre, the processor 111 may determine that at least one first genre preferred by the first user is the “action” genre, the “comedy” genre, and the “horror movie” genre.
於動作330中，各種實施例之處理器111可確認第1用戶是否連續視聽屬於與上述至少一個第1類型不同之第2類型之至少一個內容預定期間。預定期間可由系統10之管理者自由地設定。預定期間例如可為5天。例如，於第1用戶偏好之至少一個第1類型為「動作」類型及「喜劇」類型之情形時，處理器111可確認是否連續視聽屬於與上述第1類型不同之「悲傷(sad)」類型之至少一個內容預定期間。例如，處理器111可確認第1用戶是否連續5天視聽屬於第2類型即「悲傷」類型之內容。In action 330, the processor 111 of various embodiments may confirm whether the first user has continuously viewed at least one content of the second type different from the at least one first type mentioned above for a predetermined period of time. The predetermined period may be freely set by the administrator of the system 10. The predetermined period may be, for example, 5 days. For example, when at least one of the first types preferred by the first user is the "action" type and the "comedy" type, the processor 111 may confirm whether the first user has continuously viewed at least one content of the "sad" type different from the first type mentioned above for a predetermined period of time. For example, the processor 111 may confirm whether the first user has continuously viewed content of the second type, i.e., the "sad" type, for 5 consecutive days.
於動作340中，各種實施例之處理器111可響應於第1用戶連續視聽屬於上述第2類型之至少一個內容預定期間，基於第2類型來分析第1用戶之情緒狀態。若第1用戶連續視聽屬於除偏好之至少一個第1類型以外之第2類型之至少一個內容預定期間，則處理器111可判斷第1用戶之內容視聽歷史發生變化，判斷第1用戶之情緒狀態發生變化。於上述情形時，處理器111可基於第2類型來分析第1用戶之情緒狀態。即，處理器111可探測用戶之內容視聽歷史之急遽之變化及連續性而分析第1用戶的情緒狀態。In action 340, the processor 111 of various embodiments may analyze the emotional state of the first user based on the second type in response to the first user continuously viewing at least one content of the second type for a predetermined period of time. If the first user continuously views at least one content of the second type other than the preferred at least one first type for a predetermined period of time, the processor 111 may determine that the content viewing history of the first user has changed, and determine that the emotional state of the first user has changed. In the above situation, the processor 111 may analyze the emotional state of the first user based on the second type. That is, the processor 111 may detect the rapid changes and continuity of the user's content viewing history and analyze the emotional state of the first user.
例如，於第1用戶偏好之至少一個第1類型為「動作」類型及「喜劇」類型，但連續5天視聽與至少一個第1類型不同之第2類型即「悲傷」類型之情形時，處理器111可基於「悲傷」類型，分析(推斷)出第1用戶之情緒狀態處於「悲傷(sadness)或哭泣(cry)」狀態。例如，於第1用戶偏好之至少一個第1類型為「紀錄片」類型及「恐怖片」類型，但連續5天視聽屬於與該等類型不同之第2類型即「喜劇」類型之至少一個內容之情形時，處理器111可基於「喜劇」類型，分析出第1用戶之情緒狀態處於「幸福(happiness)或笑(laugh)」狀態。For example, when the first user prefers at least one of the first types of "action" and "comedy", but has watched and listened to a second type of "sadness" that is different from at least one of the first types for five consecutive days, the processor 111 can analyze (infer) that the first user's emotional state is in a "sadness or crying" state based on the "sadness" type. For example, when at least one of the first types preferred by the first user is "documentary" and "horror", but the first user has watched at least one content of a second type different from these types, namely "comedy", for five consecutive days, the processor 111 can analyze the first user's emotional state as being in a "happiness or laughter" state based on the "comedy" type.
根據各種實施例，第1用戶之情緒狀態可為預定之複數個情緒狀態中之一者。預定之複數個情緒狀態可由系統10之管理者設定。預定之複數個情緒狀態可包括「幸福(happiness)或笑(laugh)」狀態、「悲傷(sadness)或哭泣(cry)」狀態、「憤怒(rage)」狀態、「平和(calm)」狀態、「憂鬱(gloomy)」狀態、或「疲憊(tired)」狀態。上述預定之複數個情緒狀態為例示性者，可設定各種情緒狀態。複數個情緒狀態各者可與至少一個類型對應。例如，「幸福或笑」狀態可與「喜劇」類型對應。例如，「憤怒」狀態可與「動作」類型對應。According to various embodiments, the emotional state of the first user may be one of a plurality of predetermined emotional states. The plurality of predetermined emotional states may be set by the administrator of the system 10. The plurality of predetermined emotional states may include a "happiness or laughter" state, a "sadness or cry" state, a "rage" state, a "calm" state, a "gloomy" state, or a "tired" state. The plurality of predetermined emotional states are exemplary, and various emotional states may be set. Each of the plurality of emotional states may correspond to at least one type. For example, the "happiness or laughter" state may correspond to the "comedy" type. For example, the "Angry" state can be mapped to the "Action" type.
根據一實施例，處理器111除基於第1用戶之視聽歷史之變化以外，亦可基於天氣資訊來分析第1用戶之情緒狀態。處理器111可自第1用戶之終端裝置120a收集第1用戶之位置資訊。處理器111可自第2外部伺服器收集與第1用戶之位置資訊對應之天氣資訊。第2外部伺服器可為提供天氣資訊之伺服器。處理器111可基於第2類型及天氣資訊，分析第1用戶之情緒狀態。例如，於第2類型為「悲傷」類型且當前之天氣資訊為「雨(rain)」之情形時，處理器111可分析出第1用戶之情緒狀態處於「悲傷或哭泣」狀態。According to one embodiment, the processor 111 may analyze the emotional state of the first user based on weather information in addition to the changes in the viewing and audio history of the first user. The processor 111 may collect the location information of the first user from the terminal device 120a of the first user. The processor 111 may collect weather information corresponding to the location information of the first user from the second external server. The second external server may be a server that provides weather information. The processor 111 may analyze the emotional state of the first user based on the second type and the weather information. For example, when the second type is the "sad" type and the current weather information is "rain", the processor 111 may analyze that the emotional state of the first user is in a "sad or crying" state.
於動作350中，各種實施例之處理器111可基於第1用戶之情緒狀態，確定上述複數個內容中之針對上述第1用戶之推薦內容。例如，於分析出第1用戶之情緒狀態處於「悲傷或哭泣」狀態之情形時，處理器111可將適於「悲傷或哭泣」狀態之內容即悲傷電影「M」確定為推薦內容。即，於第1用戶之情緒狀態為「悲傷或哭泣」狀態之情形時，處理器111可判斷第1用戶處於想要哭泣之狀態，並將悲傷電影「M」確定為推薦內容。In action 350, the processor 111 of various embodiments may determine the recommended content for the first user among the plurality of contents based on the emotional state of the first user. For example, when analyzing that the emotional state of the first user is in the "sad or crying" state, the processor 111 may determine the content suitable for the "sad or crying" state, i.e., the sad movie "M", as the recommended content. That is, when the emotional state of the first user is in the "sad or crying" state, the processor 111 may determine that the first user is in a state of wanting to cry, and determine the sad movie "M" as the recommended content.
例如，於分析出第1用戶之情緒狀態處於「幸福或笑」狀態之情形時，處理器111可將適於「幸福或笑」狀態之內容即搞笑綜藝「N」確定為推薦內容。即，於第1用戶之情緒狀態為「幸福或笑」狀態之情形時，處理器111可判斷第1用戶處於想要笑之狀態，並將搞笑綜藝「N」確定為推薦內容。For example, when the first user's emotional state is analyzed to be in a "happy or laughing" state, the processor 111 may determine that the first user is in a "happy or laughing" state, i.e., the comedy variety show "N", as the recommended content. That is, when the first user's emotional state is in a "happy or laughing" state, the processor 111 may determine that the first user is in a state of wanting to laugh, and determine the comedy variety show "N" as the recommended content.
於動作360中，各種實施例之處理器111可向上述第1用戶之終端裝置120a傳輸確定之推薦內容。第1用戶可藉由第1用戶之終端裝置120a而接收推薦內容。In action 360, the processor 111 of various embodiments may transmit the determined recommended content to the terminal device 120a of the first user. The first user may receive the recommended content via the terminal device 120a of the first user.
各種實施例之處理器111可連同確定之推薦內容一併將第1用戶之情緒狀態之資訊傳輸至第1用戶的終端裝置120a。第1用戶可藉由第1用戶之終端裝置120a而連同推薦內容一併接收自身之情緒狀態之資訊。The processor 111 of various embodiments may transmit the information of the first user's emotional state together with the determined recommended content to the first user's terminal device 120a. The first user may receive the information of his own emotional state together with the recommended content through the first user's terminal device 120a.
圖4係本發明之各種實施例之電子裝置110之動作流程圖。FIG. 4 is a flowchart of the operation of the electronic device 110 according to various embodiments of the present invention.
參照動作流程圖400，於動作410中，各種實施例之電子裝置110之處理器111可自第1外部伺服器收集複數個內容各者之資訊。第1外部伺服器例如可為運營門戶網站之伺服器。第1外部伺服器可儲存複數個內容各者之概要、用戶之評論、及元資料。元資料例如可為內容之屬性之資訊。第1外部伺服器可儲存複數個內容各者之資訊。處理器111可藉由自第1外部伺服器爬取複數個內容各者之概要、評論及元資料，收集複數個內容各者之資訊。Referring to the action flow chart 400, in action 410, the processor 111 of the electronic device 110 of various embodiments may collect information of each of a plurality of contents from the first external server. The first external server may be, for example, a server operating a portal website. The first external server may store summaries of each of a plurality of contents, comments of users, and metadata. The metadata may be, for example, information of the attributes of the content. The first external server may store information of each of a plurality of contents. The processor 111 may collect information of each of a plurality of contents by crawling summaries, comments, and metadata of each of a plurality of contents from the first external server.
於動作420中，各種實施例之處理器111可基於複數個內容各者之資訊，確定適於複數個內容各者之複數個標籤。於動作430中，各種實施例之處理器111可對複數個內容分別賦予上述複數個標籤。例如，於複數個內容各者之資訊中之「A」電影之資訊包括「眼淚」或「痛哭」之內容的情形時，處理器111可確定適於「A」電影之標籤為「悲傷」標籤並賦予。例如，於「B」綜藝之資訊包括「笑」或「有趣」之內容之情形時，處理器111可確定適於「B」綜藝之標籤為「笑」標籤並賦予。In action 420, the processor 111 of various embodiments may determine a plurality of labels suitable for each of the plurality of contents based on the information of each of the plurality of contents. In action 430, the processor 111 of various embodiments may assign the above-mentioned plurality of labels to the plurality of contents respectively. For example, when the information of the movie "A" in the information of each of the plurality of contents includes the content of "tears" or "crying", the processor 111 may determine that the label suitable for the movie "A" is the "sad" label and assign it. For example, when the information of the variety show "B" includes the content of "laughing" or "funny", the processor 111 may determine that the label suitable for the variety show "B" is the "laughing" label and assign it.
於動作440中，各種實施例之處理器111可基於複數個標籤來確定上述複數個內容各者之類型。處理器111可基於對複數個內容各者賦予之複數個標籤，確定複數個內容各者之類型。於「A」電影被賦予「悲傷」標籤之情形時，處理器111可確定「A」電影之類型為「悲傷」類型。In action 440, the processor 111 of various embodiments may determine the type of each of the plurality of contents based on the plurality of labels. The processor 111 may determine the type of each of the plurality of contents based on the plurality of labels assigned to each of the plurality of contents. In the case where the "A" movie is assigned the "sad" label, the processor 111 may determine that the type of the "A" movie is the "sad" type.
根據一實施例，處理器111亦可藉由經學習之人工智慧模型，對複數個內容各者賦予適當之標籤。人工智慧模型可為以如下方式學習之相關模型：於輸入複數個內容各者之資訊之情形時，輸出適於複數個內容各者之複數個標籤。於下文中對人工智慧模型之具體之學習方法進行敍述。According to one embodiment, the processor 111 may also assign appropriate labels to each of the plurality of contents through a learned artificial intelligence model. The artificial intelligence model may be a correlation model learned in the following manner: when information of each of the plurality of contents is input, a plurality of labels suitable for each of the plurality of contents is output. The specific learning method of the artificial intelligence model is described below.
圖5a及圖5b係表示本發明之各種實施例之構建人工智慧模型500之方法及運用人工智慧模型500之方法的圖。FIG. 5 a and FIG. 5 b are diagrams showing methods of constructing an artificial intelligence model 500 and methods of using the artificial intelligence model 500 according to various embodiments of the present invention.
參照圖5a，各種實施例之電子裝置110之處理器111可藉由如下方式來構建人工智慧模型500：利用機器學習演算法來對輸入資料集510與輸出資料集520之間的相關關係進行建模，該輸入資料集包括複數個樣本內容各者之資訊，該輸出資料集包括適於上述複數個樣本內容各者之複數個標籤。即，人工智慧模型500可藉由如下方式來產生人工智慧模型500：將上述輸入資料集510用作學習用(訓練用)輸入資料，將上述輸出資料集520用作學習用(訓練用)輸出資料。5a, the processor 111 of the electronic device 110 of various embodiments may construct an artificial intelligence model 500 in the following manner: using a machine learning algorithm to model the correlation between an input data set 510 and an output data set 520, wherein the input data set includes information of each of a plurality of sample contents, and the output data set includes a plurality of labels suitable for each of the plurality of sample contents. That is, the artificial intelligence model 500 may be generated in the following manner: using the input data set 510 as learning (training) input data, and using the output data set 520 as learning (training) output data.
例如，特定樣本內容之資訊可包括「眼淚」或「痛哭」等資訊。適於上述特定樣本內容之標籤可為「悲傷」標籤。即，可藉由學習「眼淚」或「痛哭」之資訊之資料與「悲傷」標籤之資料之間的相關關係而構建人工智慧模型500。藉由如上所述之方法，可將複數個樣本內容各者之資訊及適於複數個樣本內容各者之複數個標籤用作學習用資料。For example, the information of a specific sample content may include information such as "tears" or "crying". The label suitable for the above-mentioned specific sample content may be the label "sadness". That is, the artificial intelligence model 500 can be constructed by learning the correlation between the data of the information of "tears" or "crying" and the data of the label "sadness". By the method described above, the information of each of the plurality of sample contents and the plurality of labels suitable for each of the plurality of sample contents can be used as learning data.
於本發明中，人工智慧模型、機器學習模型或神經網模型能夠以將人腦構造實現於電腦上之方式設計，可包括複數個網路節點，該等網路節點係模擬人類神經網之神經元(neuron)且具有加權值者。複數個網路節點可模擬神經元藉由突觸(synapse)而收發信號之神經元之突觸(synaptic)活動，從而實現彼此間之連接關係。於人工智慧學習模型中，複數個網路節點可位於不同深度之層，並且根據卷積(convolution)連接關係來收發資料。人工智慧學習模型例如可為人工神經網模型(artificial neural network)、卷積類神經網模型(convolution neural network)等。In the present invention, an artificial intelligence model, a machine learning model or a neural network model can be designed in a way that the human brain structure is realized on a computer, and can include a plurality of network nodes, which are neurons that simulate the human neural network and have weighted values. A plurality of network nodes can simulate the synaptic activities of neurons that send and receive signals through synapses, thereby realizing the connection relationship between each other. In an artificial intelligence learning model, a plurality of network nodes can be located at layers of different depths, and send and receive data according to convolution connection relationships. An artificial intelligence learning model can be, for example, an artificial neural network model, a convolution neural network model, etc.
於本發明中，機器學習演算法可指深層神經網、遞歸神經網絡、卷積類神經網、用以進行分類-回歸分析之機器學習模型或強化學習模型中之一者。In the present invention, the machine learning algorithm may refer to one of a deep neural network, a recursive neural network, a convolutional neural network, a machine learning model for classification-regression analysis, or a reinforcement learning model.
參照圖5b，如上所述，可根據機器學習演算法來構建人工智慧模型500。5b, as described above, an artificial intelligence model 500 may be constructed based on a machine learning algorithm.
各種實施例之電子裝置110之處理器111可利用構建之人工智慧模型500來獲得適於複數個內容各者的複數個標籤。處理器111可將複數個內容各者之資訊作為輸入資料530輸入至人工智慧模型500。於該情形時，人工智慧模型500可輸出適於複數個內容各者之複數個標籤作為輸出資料540。處理器111可自人工智慧模型500獲得適於複數個內容各者之複數個標籤。即，可利用構建之人工智慧模型500來準確且快速地分析適於複數個內容各者之標籤。The processor 111 of the electronic device 110 of various embodiments can use the constructed artificial intelligence model 500 to obtain a plurality of labels suitable for each of the plurality of contents. The processor 111 can input the information of each of the plurality of contents as input data 530 to the artificial intelligence model 500. In this case, the artificial intelligence model 500 can output a plurality of labels suitable for each of the plurality of contents as output data 540. The processor 111 can obtain a plurality of labels suitable for each of the plurality of contents from the artificial intelligence model 500. That is, the constructed artificial intelligence model 500 can be used to accurately and quickly analyze the labels suitable for each of the plurality of contents.
圖6係本發明之各種實施例之電子裝置110之動作流程圖。FIG. 6 is a flowchart of the operation of the electronic device 110 according to various embodiments of the present invention.
參照動作流程圖600，於動作610中，各種實施例之電子裝置110之處理器111可針對複數個內容各者提取至少一個影像。於內容為視訊內容之情形時，內容可包括複數個擷取影像或靜態剪切影像。上述影像可儲存於記憶體113中。處理器111可針對複數個內容各者提取至少一個影像。Referring to the action flow chart 600, in action 610, the processor 111 of the electronic device 110 of various embodiments may extract at least one image for each of a plurality of contents. In the case where the content is video content, the content may include a plurality of captured images or still cut images. The above images may be stored in the memory 113. The processor 111 may extract at least one image for each of a plurality of contents.
於動作620中，各種實施例之處理器111可獲得上述複數個用戶對複數個內容各者之評價資訊。複數個用戶對複數個內容各者之評價資訊可儲存於記憶體113中。複數個用戶之評價資訊可為評價內容之分數資訊或對內容之評論。In action 620, the processor 111 of various embodiments may obtain the evaluation information of the plurality of users on each of the plurality of contents. The evaluation information of the plurality of users on each of the plurality of contents may be stored in the memory 113. The evaluation information of the plurality of users may be score information of the evaluation content or comments on the content.
於動作630中，各種實施例之處理器111可基於至少一個影像及上述複數個用戶之評價資訊，確定適於複數個內容各者之複數個標籤。於動作640中，各種實施例之處理器111可對複數個內容分別賦予上述複數個標籤。處理器111例如可藉由分析至少一個影像之彩度來判斷複數個內容各者之氛圍。例如，於判斷自特定內容提取之至少一個影像之彩度較低之情形時，處理器111可確定適於特定內容之標籤為「悲傷」標籤。例如，於複數個用戶各者對特定內容之評價資訊包括「眼淚」或「痛哭」之內容之情形時，處理器111可確定適於特定內容之標籤為「悲傷」標籤。In action 630, the processor 111 of various embodiments may determine a plurality of labels suitable for each of the plurality of contents based on at least one image and the evaluation information of the plurality of users. In action 640, the processor 111 of various embodiments may assign the plurality of labels to the plurality of contents respectively. The processor 111 may, for example, determine the atmosphere of each of the plurality of contents by analyzing the chromaticity of at least one image. For example, when it is determined that the chromaticity of at least one image extracted from a specific content is low, the processor 111 may determine that the label suitable for the specific content is a "sad" label. For example, when the evaluation information of multiple users on specific content includes "tears" or "crying", the processor 111 may determine that the label suitable for the specific content is the "sad" label.
於動作650中，各種實施例之處理器111可基於複數個標籤，確定上述複數個內容各者之類型。In action 650, the processor 111 of various embodiments may determine the type of each of the plurality of contents based on the plurality of tags.
於圖式所示之流程圖中，依序對程序步驟、方法步驟、演算法等進行了說明，但該等程序、方法及演算法能夠以按照任意適合之順序來進行操作之方式構成。換言之，於本發明之各種實施例中說明之程序、方法及演算法之步驟不必按照本發明中記述之順序實行。又，即便一部分步驟被說明為非同時實行，於其他實施例中亦可同時實行此種一部分步驟。又，圖式中所描述之程序之示例並不意味著所例示之程序排除對其之其他變化及修改，並不意味著所例示之程序或其步驟中之任意一者對於本發明之各種實施例中之一個以上之實施例而言為必不可少者，且並不意味著所例示之程序為較佳之程序。In the flowchart shown in the figure, the program steps, method steps, algorithms, etc. are explained in sequence, but the programs, methods, and algorithms can be constructed in a manner that the operations are performed in any suitable order. In other words, the steps of the programs, methods, and algorithms described in the various embodiments of the present invention do not have to be implemented in the order described in the present invention. Moreover, even if a part of the steps is described as not being implemented simultaneously, such a part of the steps can also be implemented simultaneously in other embodiments. Moreover, the examples of the programs described in the figures do not mean that the exemplified programs exclude other changes and modifications thereto, nor do they mean that the exemplified programs or any one of their steps are indispensable for one or more of the various embodiments of the present invention, and they do not mean that the exemplified programs are preferred programs.
藉由特定實施例而對上述方法進行了說明，但上述方法亦可於電腦可讀記錄媒體中實現為電腦可讀代碼。電腦可讀記錄媒體可包括儲存能夠由電腦系統讀取之資料之所有種類之記錄裝置。作為一例，電腦可讀記錄媒體可包括ROM(Read only memory，唯讀記憶體)、RAM(Random-access memory，隨機存取記憶體)、CD-ROM(Compact Disc Read-Only Memory，光碟唯讀記憶體)、磁帶、軟碟、光學資料儲存裝置等。又，電腦可讀記錄媒體分散於連接於網路之電腦系統，從而能夠以分散方式儲存電腦可讀代碼並執行。另外，本發明所屬之技術領域內之程式設計師可容易地推測出用以實現上述實施例之功能(functional)程式、代碼及代碼段。The above method is described by a specific embodiment, but the above method can also be implemented as a computer-readable code in a computer-readable recording medium. The computer-readable recording medium may include all kinds of recording devices that store data that can be read by a computer system. As an example, the computer-readable recording medium may include ROM (Read only memory), RAM (Random-access memory), CD-ROM (Compact Disc Read-Only Memory), magnetic tape, floppy disk, optical data storage device, etc. In addition, the computer-readable recording medium is distributed in a computer system connected to a network, so that the computer-readable code can be stored and executed in a distributed manner. In addition, programmers within the technical field to which the present invention belongs can easily infer functional programs, codes, and code segments for implementing the above-mentioned embodiments.
10: 系統 110: 電子裝置 111: 處理器 113: 記憶體 115: 通訊電路 120: 終端裝置 120a: 終端裝置 120b: 終端裝置 120c: 終端裝置 121: 控制器 123: 記憶體 125: 顯示器 127: 通訊電路 300: 動作流程圖 310: 動作 320: 動作 330: 動作 340: 動作 350: 動作 360: 動作 400: 動作流程圖 410: 動作 420: 動作 430: 動作 440: 動作 500: 人工智慧模型 510: 輸入資料集 520: 輸出資料集 530: 輸入資料 540: 輸出資料 600: 動作流程圖 610: 動作 620: 動作 630: 動作 640: 動作 650: 動作10: System110: Electronic device111: Processor113: Memory115: Communication circuit120: Terminal device120a: Terminal device120b: Terminal device120c: Terminal device121: Controller123: Memory125: Display127: Communication circuit300: Action flow chart310: Action320: Action330: Action340: Action350: Action360: Action400: Action flow chart410: Action420: Action430: Action440: Action500: Artificial Intelligence Model510: Input Dataset520: Output Dataset530: Input Data540: Output Data600: Action Flowchart610: Action620: Action630: Action640: Action650: Action
圖1係表示本發明之各種實施例之系統之圖。 圖2係表示本發明之各種實施例之電子裝置及終端裝置之方塊圖。 圖3係本發明之各種實施例之電子裝置之動作流程圖。 圖4係本發明之各種實施例之電子裝置之動作流程圖。 圖5a及圖5b係表示本發明之各種實施例之構建人工智慧模型之方法及運用人工智慧模型之方法的圖。 圖6係本發明之各種實施例之電子裝置之動作流程圖。FIG. 1 is a diagram showing a system of various embodiments of the present invention.FIG. 2 is a block diagram showing an electronic device and a terminal device of various embodiments of the present invention.FIG. 3 is a flowchart of an operation of an electronic device of various embodiments of the present invention.FIG. 4 is a flowchart of an operation of an electronic device of various embodiments of the present invention.FIG. 5a and FIG. 5b are diagrams showing a method of constructing an artificial intelligence model and a method of using an artificial intelligence model of various embodiments of the present invention.FIG. 6 is a flowchart of an operation of an electronic device of various embodiments of the present invention.
10: 系統 110: 電子裝置 120: 終端裝置 120a: 終端裝置 120b: 終端裝置 120c: 終端裝置10: System110: Electronic device120: Terminal device120a: Terminal device120b: Terminal device120c: Terminal device
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020220186076AKR102617004B1 (en) | 2022-12-27 | 2022-12-27 | Electronic device and method of providing recommended contents |
| KR10-2022-0186076 | 2022-12-27 |
| Publication Number | Publication Date |
|---|---|
| TW202433940A TW202433940A (en) | 2024-08-16 |
| TWI860182Btrue TWI860182B (en) | 2024-10-21 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW113139435ATW202508291A (en) | 2022-12-27 | 2023-11-29 | Electronic device and method of providing recommended contents |
| TW112146220ATWI860182B (en) | 2022-12-27 | 2023-11-29 | Electronic device and method of providing recommended contents |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW113139435ATW202508291A (en) | 2022-12-27 | 2023-11-29 | Electronic device and method of providing recommended contents |
| Country | Link |
|---|---|
| KR (2) | KR102617004B1 (en) |
| TW (2) | TW202508291A (en) |
| WO (1) | WO2024143664A1 (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9613033B2 (en)* | 2014-08-29 | 2017-04-04 | Yahoo!, Inc. | Emotionally relevant content |
| US20200021886A1 (en)* | 2019-08-26 | 2020-01-16 | Lg Electronics Inc. | System, apparatus and method for providing services based on preferences |
| US20210248656A1 (en)* | 2019-10-30 | 2021-08-12 | Lululemon Athletica Canada Inc. | Method and system for an interface for personalization or recommendation of products |
| US20220132214A1 (en)* | 2017-12-22 | 2022-04-28 | Hillel Felman | Systems and Methods for Annotating Video Media with Shared, Time-Synchronized, Personal Reactions |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110052997A (en)* | 2009-11-13 | 2011-05-19 | 삼성전자주식회사 | Display device and client terminal and control method thereof |
| KR20180070471A (en)* | 2016-12-16 | 2018-06-26 | 김유식 | Apparatus and method for sharing contents |
| KR102378412B1 (en)* | 2020-09-01 | 2022-03-25 | 주식회사 인생노트에이아이 | A Weather-based emotion index provision method and a device therefor |
| KR20220085219A (en)* | 2020-12-15 | 2022-06-22 | 주식회사 와이즈넛 | The method of filtering ads using youtube video metadata |
| KR102339377B1 (en)* | 2021-03-29 | 2021-12-13 | 김지혜 | Method and device for providing customized cultural and art contents |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9613033B2 (en)* | 2014-08-29 | 2017-04-04 | Yahoo!, Inc. | Emotionally relevant content |
| US20220132214A1 (en)* | 2017-12-22 | 2022-04-28 | Hillel Felman | Systems and Methods for Annotating Video Media with Shared, Time-Synchronized, Personal Reactions |
| US20200021886A1 (en)* | 2019-08-26 | 2020-01-16 | Lg Electronics Inc. | System, apparatus and method for providing services based on preferences |
| US20210248656A1 (en)* | 2019-10-30 | 2021-08-12 | Lululemon Athletica Canada Inc. | Method and system for an interface for personalization or recommendation of products |
| Publication number | Publication date |
|---|---|
| TW202508291A (en) | 2025-02-16 |
| KR102617004B1 (en) | 2023-12-27 |
| TW202433940A (en) | 2024-08-16 |
| WO2024143664A1 (en) | 2024-07-04 |
| KR20240103991A (en) | 2024-07-04 |
| KR102855783B1 (en) | 2025-09-08 |
| Publication | Publication Date | Title |
|---|---|---|
| CN111444428B (en) | Information recommendation method and device based on artificial intelligence, electronic equipment and storage medium | |
| CN110569377B (en) | Media file processing method and device | |
| CN113901337B (en) | Application recommendation method, device, equipment and computer readable storage medium | |
| CN117009650A (en) | Recommendation method and device | |
| US11914657B2 (en) | Machine learning aided automatic taxonomy for web data | |
| CN106874936A (en) | Image propagates monitoring method and device | |
| CN115618024A (en) | Multimedia recommendation method and device and electronic equipment | |
| CN116205686A (en) | Method, device, equipment and storage medium for recommending multimedia resources | |
| Pichandi et al. | Hybrid deep models for parallel feature extraction and enhanced emotion state classification | |
| Bokhare et al. | Emotion detection-based video recommendation system using machine learning and deep learning framework | |
| CN113326760A (en) | Video classification method and device | |
| Shome et al. | A generalized mechanism beyond NLP for real-time detection of cyber abuse through facial expression analytics | |
| CN116595252A (en) | A data processing method and related device | |
| Sharma et al. | A survey and classification on recommendation systems | |
| Shoba et al. | Aging facial recognition for feature extraction using adaptive fully recurrent deep neural learning | |
| CN120014648A (en) | Video resource representation method, coding model training method and device | |
| CN118260380B (en) | Processing method and system for multimedia scene interaction data | |
| TWI860182B (en) | Electronic device and method of providing recommended contents | |
| Zhu | Visual contextual perception and user emotional feedback in visual communication design | |
| Skaka‐Čekić et al. | A feature selection for video quality of experience modeling: A systematic literature review | |
| Alabdulrahman et al. | Active learning and deep learning for the cold-start problem in recommendation system: A comparative study | |
| Siddiqi et al. | Evaluating real-life performance of the state-of-the-art in facial expression recognition using a novel YouTube-based datasets | |
| CN116543798A (en) | Emotion recognition method and device based on multiple classifiers, electronic equipment and medium | |
| TWI871788B (en) | Electronic device and method of providing recommended contents | |
| CN115204928A (en) | User data processing method, device, computer equipment and storage medium |