1 |
社群網路上的城市情感聚合表現與觀察 / Observation about World Cities’ Emotions with Online Social Communities張伸吉, Chang, Shen Chi Unknown Date (has links)
隨著社群網站的大量崛起,人們在網路上的行為從早期的單方面獲得資訊,慢慢轉變到與他人的網頁互動,以至於到最近的發展成社群(Community)關係。網路使用者對虛擬社群的好奇,導致了相關社群網站的蓬勃發展。近年來對於社群網路的研究日漸興盛,各個研究領域都分別以其擅長的角度嘗試切入,想了解為何社群網路如此快速的崛起、社群網路形成的各種架構或是社群網路上的大量資料呈現的資訊以及其分析等等。
社群網站滿足了人們期待與其他人產生互動、情感維繫以及得到更多資訊的需要,提供了一個虛擬空間,讓關心相同主題的使用者群聚在一起並且分享資訊。無論是想與他人互動,或是情感維繫以及渴望得到資訊,這些動機皆與人的情感表現息息相關,本研究即是採用情感的角度,來觀察社群中的特定行動者與其網路。
有關世界城市網路的研究目前並不多見,本研究嘗試以城市為社會網路中的行動者角色,研究一個由網路相片分享構築成的特殊社群網站Flickr。利用相片可忠實呈現拍攝者與被拍攝者情感的特性,以「城市」與「情感」兩大基礎來觀察此社群,嘗試構築出一個有關世界城市情感面向的網路關係。根據本研究目的,我們將建立「城市-情感共現網路」,來發掘世界主要城市之間隱含的連結與關係,或是其隱藏的情感表現。 / Along with considerable growth of social network websites, people’s actions on internet changed slowly from acquiring information to exchanging information and interacting with other people via web pages, and eventually this change has created a so-called “community relationship” on line. This blossom of relevant social network websites resulted from internet users’ curiosity about the invented virtual community. Numerous researchers in each relevant field have donated themselves into analysis and researches aiming to understand the reasons why all kinds of community websites have been created so quickly, how these communities have been structured, how the information and data underlying these websites have been presented and analyzed.
The goal of this paper is to dig out links and relations between big cities worldwide and human emotions in these cities via community websites. We tried to analyze a unique community website “Flickr” that functions on the base of uploading pictures only. We categorized the pictures on Flickr on two pillar axes, “Region” and “Emotion”, as the emotions of photo shooters and personages have been detailed and recorded by these pictures. By the above categories, We drew a map of emotions in cities that will reveal a 2-mode network of emotions and cities.
|
2 |
Flickr網站上世界商務城市之情感輪廓 / Emotional Contours of the Commerce Cities on the Website Flickr馮成發, Fong, Chen Fa Unknown Date (has links)
近年來電腦科學的進步只能以一日千里來形容,不管在軟體或是硬體方面都有驚人的發展,軟體方面有網際網路Web 2.0技術的興盛及普及,使得人們在分享及交流資訊更加快速且便利,硬體方面則有數位相機和有照相功能智慧型手機的發明,造就了分享資訊很快的從文字模式演變成影音、相片等多媒體模式。Flickr社群網站為目前網路世界裡最重要的相片分享平台,每個人都可以將生活中擁有喜、怒、哀、樂情緒的相片上傳至該網站上與他人分享,而且此網站平台也提供下標籤功能,讓上傳者可以更正確的傳達要分享的情感。如當相片被加註上快樂的標籤,也就代表上傳者對這張相片當時的環境情緒反應為愉快、或甚至於興奮,相反地;當相片被加註上生氣的標籤,就表示該相片給上傳者的情緒反應是不愉快的、或甚至於憤怒。當同一區域(如城市)透過大量情感標籤的累積,自然而然就會呈現出該區域的情感輪廓。
情緒議題的研究近年來在各知識領域中已被廣泛的討論著,但針對區域性的情緒表現之研究探討似乎還不多。本研究藉由Flickr社群網站的全球性特質,結合Derudder and Taylor兩位學者於2005年提出的「The cliquishness of world cities」研究報告,定義出41個商務活動頻繁城市作為本研究的研究範圍,並應用Flickr社群網站上強大又完整的API介面功能,撰寫Client端程式擷取這些城市在Flickr網站上有加註情緒標籤的相片數共761,854張、其相關的標籤數有21,569,593個,再經由本研究提出的研究方法及步驟,逐一處理這些各城市相片上傳者所加註的大量標籤,就可以找出每個城市各情感象限數量最多的前30個標籤當作顯著標籤。
最後本研究綜合分析從Flickr網站上取得的大量城市、相片、及顯著標籤相關資料,分別計算出每個城市正負向情感象限的強度百分比,再以正向情感象限強度百分比為基準,定義出這些商務活動頻繁城市的「快樂指數」數值;並利用社會網絡分析軟體NodeXL來觀察各城市、情感性標籤與顯著標籤所呈現的網絡關係。 / In recent years, the computer science progress is extremely fast, whether in software or hardware has an alarming growth. The software aspect has the Internet Web 2.0 technology prosperity and popular, causes the people in share and exchange information are faster and convenient. The hardware aspect has the digital cameras and the smartphones invention, causes the share information from the writing pattern to the multimedia patterns very quickly. The Flickr social website is the most important of shared photograph in the network world for currently,everyone can shared the joy, anger, sadness, happy mood photograph by uploading to this website. This website platform also provides the tagging function, lets the uploader can more correct transmission their emotion. When the identical region (such as a city) through a large number of emotional labels cumulatively, naturally will be showing the emotional contours of the region.
Emotional issues have been widespread discussion in various area of knowledge in recent years, but research the performance of emotion for region seems not much. This research because of Flickr social website global special characteristic, combined Derudder and Taylor two scholars to propose "The cliquishness of world cities" research reports in 2005, Defines 41 economics and trade activity frequent city to take this research the study scope.
Finally, this research made a comprehensive analysis by a large number of cities, photos, and significant label information from the Flickr website, and calculates the percentage of each city to the strength of positive and negative emotions quadrant.Then the percentage of positive emotional intensity as a benchmark quadrant, Defines these economics and trade activity frequent city's "happiness index".
|
3 |
以社群標籤組為基礎之不同角度文章之推薦 / Using social tags for comprehensive document recommendation鄭挺拔, Cheng, Ting Pa Unknown Date (has links)
近年來,推薦系統(recommendation system)相關研究是一個很熱門的議題,當使用者看到一篇文章,對該文章所描述的事件很感興趣,想要了解該事件的全貌,此時想要得到是該事件的通盤的見解,而非局部的意見,也就是以不同角度去解析此事件的文章清單時,若以過去傳統推薦系統的作法,推薦與這篇文章相似的文章給使用者就未必合適,因為相似文章只能反映對此事件相同角度,而非對此事件不同角度的文章。因此,本研究擬使用社群性標籤(social tag)解決以上問題。透過不同使用者標註標籤反映不同看法的機制,我們可以從文章中選出代表性的標籤,透過該標籤組與文章分數計算,找出對此事件不同角度的文章清單推薦給使用者。實驗結果顯示,若文章有較高的可信度擁有多種角度,則使用我們提出的演算法確實擁有較好的準確度。
|
4 |
以Web2.0民眾分類法建置音樂推薦系統之研究 / A Music Recommendation System Based on the Web 2.0 Folksonomy Approach鄭學侖, Cheng,Allen Unknown Date (has links)
近年來,數位格式的音樂使得音樂市場活動逐漸由實體轉移到線上,消費者也開始會透過線上服務自己搜尋並取得在網路上大量的音樂。但是由於過量的音樂資訊,使得消費者在下載音樂試聽後,往往真正會去購買的比例是微乎極微,因此造成唱片業者對於音樂下載的觀點仍非常保守。因此,如何去提升在消費者下載之歌曲數量與真正消費之音樂的比例,將是線上音樂市場的一項發展重點。
本研究希望透過近年在Web 2.0網站上常見之標籤系統,實作一個由群眾定義音樂分類的音樂資訊交流平台,並基於此標籤式的分類法,發展一套推薦系統,來提高消費者接觸到喜歡之音樂的比例,近一步解決上述之問題。在系統發展中,本研究提出一套用於推薦系統之演算法則,並在建置之實驗音樂資訊交流平台上驗證其可行性。最後,本研究亦針對未來研究議題方向,提出一些建議。
|
5 |
多語系個人飲食攝影註記典藏系統輔以群眾外包 / Multilingual Personal Dietary Photograph Annotation System with The Assistance of Crowdsourcing林睦叡, Lin, Mu Rui Unknown Date (has links)
本研究於個人數位註釋應用程式iPARIS上,建立影像標籤註釋之功能,稱為iPARIS-PLUS。它提供不同於以往文字註釋的新方法,讓使用者可以有另一種選擇,也同時解決在面臨多國語系時的註釋問題,並有效的降低記錄所花費之時間。iPARIS-Plus能讓使用者保有在行動裝置上紀實之便利性的同時,也能兼顧記錄的完整性,讓人們不再將記錄視為一種麻煩。除此之外,我們透過群眾外包的力量將用於註釋的影像標籤轉換為文字後儲存於資料庫中,解決原先因多國語系註釋問題讓使用者無法輸入文字,導致資料庫缺少該筆資料而造成資料空缺。在評估方面,受測者認為影像標籤註釋之方法可以有效的解決多國語系註釋之問題,以及有效節省在行動裝置上打字之時間,更加強了記錄的便利性與完整性,同時也帶來不同以往的新鮮感。而我們藉由群眾外包得到良好的解析率,並且從歷程記錄中發現群眾外包於運作上,越多專業之群眾並不一定帶來越好的成果,只仰賴少部分專業之群眾提供貢獻,反而能減少問題產生,進而得到較好之結果。 / In this study, we created the function of image tags annotating in the application, iPARIS-Plus. It provided a new method of annotation which is different from the text annotations, therefore, users could have another choice. This function could solve the problem of multilingual annotation and reduce the time effectively when users take for the record. iPARIS-Plus allows users to retain the convenience of recording on their mobile device, at the same time, it also considers the integrity of the records, so let people will no longer feel recording is a trouble. In addition, we converted the image tags that used to annotate into text through the crowdsourcing system to solve the problem which users couldn’t enter text because of the multilingual annotation, it resulted in a lack of databases. In the evaluation, users argued that the image tags annotation method could solve the problem of multilingual annotation effectively, as well as saving the time they typing on their moblie devices, even more it can enhance the integrity and convenience of records. We got a good resolution rate of converting the image tags into text by crowdsourcing system and found that more professional crowds do not bring better results. On the contrary, we could rely on a few of professional crowds to reduce the problems, then got a better results.
|
6 |
新光電科技的創新事業經營模式 / The Business Investment Strategy of A Novel Optoelectronics Technology 王昱勝 Unknown Date (has links)
新光電科技的出現總是讓人感到異常興奮,然而由實驗室技術出發到商品化是一 段非常艱辛、漫長且危險的過程,稍有不慎就會墜入失敗的深淵中。本論文首先介紹 該技術的特點,一種可以將室內光源轉化成電力輸出的光伏科技,並且描述了該技術 在商業化時可能面臨的挑戰。全世界專利的分析顯現出該領域目前尚未有特定企業進 行全球化的布局,可以想見此技術正處於成長期並且充滿了機會。我們的個案研究選 擇台灣第一家鑽研該領域的前創有限公司進行分析,包含競爭者分析,競爭力分析與 各式各樣可能的的市場應用。在考慮到此技術的目前的優勢與劣勢,我們規劃出馬上 可以獲利的市場應用,也就是電子標籤。透過關鍵廠商的訪談使我們夠清楚知道短期 之內該將有限的資源集中在那些地方以滿足客戶的需求。營收來源根據合作模式的不 同,合作時間的長短而分成不同的獲利方式。業務推廣的模式勾勒出短、中、長期的 營運藍圖。帶領客戶一起經歷一場關於價值的創新歷程。
|
7 |
基於文件相似度的標籤推薦-應用於問答型網站 / Applying Tag Recommendation base on Document Similarity in Question and Answer Website葉早彬, Tsao, Pin Yeh Unknown Date (has links)
隨著人們習慣的改變,從網路上獲取新知漸漸取代傳統媒體,這也延伸產生許多新的行為。社群標籤是近幾年流行的一種透過使用者標記來分類與詮釋資訊的方式,相較於傳統分類學要求物件被分類到預先定義好的類別,社群標籤則沒有這樣的要求,因此容易因應內容的變動做出調整。
問答型網站是近年來興起的一種個開放性的知識分享平台,例如quora、Stack Overflow、yahoo 奇摩知識+,使用者可以在平台上與網友做問答的互動,在問與答的討論中,結合大眾的經驗與專長,幫助使用者找到滿意的答案,使用單純的問答系統的好處是可以不必在不同且以分類為主的論壇花費時間尋找答案,和在關鍵字搜索中的結果花費時間尋找答案。
本研究希望能針對問答型網站的文件做自動標籤分類,運用標籤推薦技術來幫助使用者能夠更有效率的找到需要的問題,也讓問答平台可以把這些由使用者所產生的大量問題分群歸類。
在研究過程蒐集Stack Exchange問答網站共20638個問題,使用naïve Bayes演算法與文件相似度計算的方式,進行標籤推薦,推薦適合的標籤給新進文件。在研究結果中,推薦標籤的準確率有64.2%
本研究希望透過自動分類標籤,有效地分類問題。幫助使用者有效率的找到需要的問題,也能把這些由使用者所產生的大量問題分群歸類。 / With User's behavior change. User access to new knowledge from the internet instead of from the traditional media. This Change leads to a lot new behavior. Social tagging is popular in recent years through a user tag to classify and annotate information. Unlike traditional taxonomy requiring items are classified into predefined categories, Social tagging is more elastic to adjust through the content change.
Q & A Website is the rise in recent years. Like Quora , Stack Overflow , yahoo Knowledge plus. User can interact with other people form this platform , in Q & A discussion, with People's experience and expertise to help the user find a satisfactory answer.
This study hopes to build a tag recommendation system for Q & A Website. The recommendation system can help people find the right problem efficiently , and let Q & A platform can put these numerous problems into the right place.
We collect 20,638 questions from Stack Exchange. Use naïve Bayes algorithm and document similarity calculation to recommend tag for the new document. The result of the evaluation show we can effectively recommend relevant tags for the new question.
|
8 |
應用主題探勘與標籤聚合於標籤推薦之研究 / Application of topic mining and tag clustering for tag recommendation高挺桂, Kao, Ting Kuei Unknown Date (has links)
標記社群標籤是Web2.0以來流行的一種透過使用者詮釋和分享資訊的方式,作為傳統分類方法的替代,其方便、靈活的特色使得使用者能夠輕易地因應內容標註標籤。不過其也有缺點,除了有相當多無標籤標註的內容,也存在大量模糊、不精確的標籤,降低了系統本身組織分類標籤的能力。為了解決上述兩項問題,本研究提出了一種結合主題探勘與標籤聚合的自動化標籤推薦方法,期望能夠建立一個去人工過程的自動化標籤推薦規則,來推薦合適的標籤給使用者。
本研究蒐集了痞客邦部落格中,點閱次數大於5000次的熱門中文文章共2500篇,經過前處理,並以其中1939篇訓練模型及400篇作為測試語料來驗證方法。在主題探勘部分,本研究利用LDA主題模型計算不同文章的主題語意,來與既有標籤作出關聯,而能夠針對新進文章預測主題並推薦主題相關標籤給它。其中,本研究利用了能評斷模型表現情形的混淆度(Perplexity)來協助選取LDA的主題數,改善了LDA需要人主觀決定主題數的問題;在標籤聚合部分,本研究以階層式分群法,將有共同出現過的標籤群聚起來,以便找出有相似語意概念的標籤。其中,本研究將分群停止條件設定為共現次數最少為1次,改善了分群方法需要設定分群數量才能有結果的問題,也使本方法能夠自動化的找出合適的分群數目。
實驗結果顯示,依照文章主題語意來推薦標籤有一定程度的可行性,且以混淆度所協助選取的主題數取得一致性較好的結果。而依照階層式分群所分出的標籤群中,同一群中的標籤確實擁有相似、類似的概念語意。最後,在結合主題探勘與標籤聚合的方法上,其Top-1至Top-5的準確率平均提升了14.1%,且Top-1準確率也達到72.25%。代表本研究針對文章寫作及標記標籤的習性切入的做法,確實能幫助提升標籤推薦的準確率,也代表本研究確實建立了一個自動化的標籤推薦規則,能推薦出合適的標籤來幫助使用者在撰寫文章後,能夠更方便、精確的標上標籤。 / Tags are a popular way of interpreting and sharing information through use, and as a substitute for traditional classification methods, the convenience and flexibility of the community makes it easy for users to use. But it also has disadvantages, in addition to a considerable number of non-tagged content, there are also many fuzzy and inaccurate tags. To solve these two problems, this study proposes a tag recommendation method that combines the Topic Mining and Tag Clustering.
In this study, we collected a total of 2500 articles by Pixnet as a corpus. In the Topic Mining section, this study uses the LDA Model to calculate the subject semantics of different articles to associate with existing tags, and we can predict topics for new articles to recommend topics related tags to them. Among them, the topics number of the LDA Model uses the Perplexity to help the selection. In the Tag Clustering section, this study uses the Hierarchical Clustering to collect the tags that have appeared together to find similar semantic concepts. The stop condition is set to a minimum of 1 co-occurrence times, which solves the problem that the clustering method needs to set the number of groups to have the result.
First, the Topic Mining results show that it is feasible to recommend tags according to the semantics of the article, and the experiment proves that the number of topics chosen according to the Perplexity is superior to the other topics. Second, the Tag Clustering results show that the same group of tags does have similar conceptual semantics. Last, experiments show that the accuracy rate of Top-1 to Top-5 in combination with two methods increased average of 14.1%, and its Top-1 accuracy rate is 72.25%,and it tells that our tag recommendation method can recommend the appropriate tag for users to use.
|
9 |
在社會網路上透過Tag-Thesaurus模型達到有效的資源彙整 / Resource Aggregation via Tag-Thesaurus model on Social Web宋昆銘 Unknown Date (has links)
我們從自然語言領域中借用Thesaurus模型作為字彙關聯的基礎,陸續加入Folksonomy概念、Social Network Service指標的蒐集以及domain-specific ontology來建構Tag-Thesaurus模型,用來解決使用一般tagging system資源彙整能力不足的問題。首先我們對將要實驗的領域選取初始字彙,並利用這些字彙建構Tag-Thesaurus模型。接著將預先準備的這些字彙釋放到社會網路服務平台的tagging system中,透過社會網路服務平台中的tagging system來蒐集使用者對於資源的平面分類資訊,利用這些資訊來對Tag-Thesaurus模型持續地擴充。透過這樣的Tag-Thesaurus模型,我們將可以獲得較佳的資源彙整。domain-specific ontology的加入將可以強化由上而下的資源彙整。而Social Network Service當中的其他資訊,如FOAF[16]或是個人的偏好等,將可以提昇個人化資源彙整的能力。這樣的結合方式不僅是ontology應用的示範,我們更希望透過這樣的混合式模型,使得Web 2.0這樣子廣泛蒐集眾人智慧的概念能夠成為跨入語意網的橋樑。 / We aggregate various resources through the Tag-Thesaurus Model. There are three parts in Tag-Thesaurus model, the Folksonomy formal model, indices collection on Social Network Service, and lightweight domain-specific ontology. The Folksnomy model reconstruct relationships between tags, and we can aggregate resources by tags. The indices collection on Social Network Service help us to decide which resource are more important. Finally, the lightweight domain-specific ontology provide the standard interface to describe the relationships between tags.
|
10 |
利用標籤社會網絡之影響力最大化達到目標式廣告行銷 / Influence maximization in labeled social network for target advertising李法賢 Unknown Date (has links)
病毒式行銷(viral marketing)透過人際之間的互動,藉由消費者對消費者的推薦,達到廣告效益。而廣告商要如何進行病毒式行銷?廣告商必須在有限資源下從人群中找出具有影響力的人,將產品或是概念推薦給更多的消費者,以說服消費者會採納他們的意見。
利用社會網絡(Social network),我們可以簡單地將消費者之間的關係用節點跟邊呈現,而Influence Maximization就是在社會網絡上選擇最具有影響力的k個消費者,而這k個消費者能影響最多的消費者。
然而,廣告相當重視目標消費群,廣告目的就是希望能夠影響目標消費群,使目標消費群購買產品。因此,我們針對標籤社會網絡(Labeled social network)提出Labeled influence maximization的問題,不同過往的研究,我們加入了標籤的條件,希望在標籤社會網絡中影響到最多符合標籤條件的節點。
針對標籤社會網絡,我們除了修改四個解決Influence maximization problem的方法,Greedy、NewGreedy、CELFGreedy和DegreeDiscount,以找出影響最多符合類別條件的節點的趨近解。我們也提出了兩個新的方法ProximityDiscount和MaximumCoverage來解決Labeled influence maximization problem。我們在Offline時,先計算節點與節點之間的Proximity,當行銷人員Online擬定行效策略時,系統利用Proximity,Onlin找出趨近解。
實驗所採用的資料是Internet Movie Database的社會網絡。根據實驗結果顯示,在兼顧效率與效果的情況下,適合用ProximityDiscount來解決Labeled influence maximization problem。 / Influence maximization problem is to find a small subset of nodes (seed nodes) in a social network that could maximize the spread of influence. But when marketers advertise for some products, they have a set of target audience. However, influence maximization doesn’t take target audience into account.
This thesis addresses a new problem called labeled influence maximization problem, which is to find a subset of nodes in a labeled social network that could influence target audience and maximizes the profit of influence. In labeled social network, every node has a label, and every label has profit which can be set by marketers.
We propose six algorithms to solve labeled influence maximization problem. To accommodate the objective of labeled influence maximization, four algorithms, called LabeledGreedy, LabeledNewGreedy, LabeledCELFGreedy, and LabeledDegreeDiscount, are modified from previous studies on original influence maximization. Moreover, we propose two new algorithms, called ProximityDiscount and MaximumCoverage, which offline compute the proximities of any two nodes in the labeled social network. When marketers make strategies online, the system will return the approximate solution by using proximities.
Experiments were performed on the labeled social network constructed from Internet Movie Database, the result shows that ProximityDiscount may provide good efficiency and effectiveness.
|
Page generated in 0.0289 seconds