講者

頁數: 1 2 3 4 5 6 7 - 每頁 20 筆

共有 131 位講者

人工智慧╳資料科學演講:

周世恩

周世恩 Demographic prediction based on public social network behaviors

年齡性別統計資訊在市場調查扮演很重要資訊,但由於隱私權規範,這些資料很難直接向使用者取得,QSearch 研發團隊透過臉書公開按讚紀錄預測與臉書公開資料年齡性別,舉例來說,可能關注美妝相關粉絲團可能女性使用者偏多,此外,我們在雲端服務上建構預測系統,能在幾分鐘內根據互動紀錄分析超過 500 萬個使用者年齡性別資料,大幅加速分析效率,同時能提供更豐富的市場調查報告。
more
周志成

周志成 Learn Data Science by Doing a Kaggle Competition

Once sequenced, a cancer tumor can have thousands of genetic mutations. But the challenge is to distinguish the mutations that contribute to tumor growth from other neutral mutations. Currently the interpretation of genetic mutations is done manually, which is time-consuming and knowledge-demanding. Therefore, Classifying Clinically Actionable Genetic Mutations challenged the Kaggle community to develop algorithms that automatically classify genetic variations based on evidence from text-based clinical literature. As a problem of natural language processing (NLP) and machine learning, this Kaggle competition is not a trivial task. The main difficulties are three fold. First of all, interpreting clinical evidence from literature is very challenging even for human specialists, since it takes expertise of domain knowledge and lengthy time of reading to understand key information in the literature and make classification accordingly. Secondly, only 3321 training data is given, which is far less compared with other Kaggle challenges and will increase the risk of overfitting. Moreover, much of the test data is machine-generated, which boosts the complexities of this task. To tackle this challenge, cooperation between teams of data science and clinical medicine was built up. Algorithms with insights from both fields were developed. To extract the key information of genetic mutation from text, hand crafted feature engineering was done with several state-of-the-art NLP methods. To obtain effective representations of the texts, we also consulted medical experts about how specialists read and classify genetic mutations, and adjusted our approaches accordingly. Furthermore, efforts were spent on classifiers to prevent issue of overfitting resulting from small training dataset. From the experience of participating in this competition, we demonstrated how cooperation between different expertise can bring in further insights to deal with challenging data science problem as this one.
more
周培廉

周培廉 我在 Kaggle 數海獅

kaggle數海獅競賽簡介、心得。使用深度學習方法協助NOAA生物學家監控阿留申群島的海獅數量。
more
周恩誌

周恩誌 以深度學習估算車牌競標成交價

764/5000 因著社會迷信,擁有吉祥數字的車輛牌照在拍賣會中往往能賣得非常高的價錢。 與其他常見的拍賣不同的是,現時車輛牌照在拍賣前並沒有估價。是項研究的目標是構建一個準確的模型用來估算香港車牌的拍賣成交價。因為車牌的價值取決於牌上的數字及其語義,車牌競標成交價的估算可以被看成為自然語言處理的一種,並就此構建相應的深度神經網絡模型。
more
周曼如

周曼如 The Journey To Become A Data Scientist-From School To Industry

What will happen to a young data scientist who just graduated from school after entering an industry? What is the difference between experiments and applications to solve real world problems using data science? In this talk, I will share my journey from a new machine learning and data mining graduated student to a senior data engineer (Or sometimes, the data scientist). How to detect and define problems? How to understand the domain knowledge and combine it into data models? How to apply data science to solve them? How to ensure the data correctness during ETL process? How to evaluate the performance? Is there any better solution? Etc. Becoming a data science unicorn for one single person is difficult. How about a group of talents? To build a successful data product, cooperation with people from different functions is needed. In the end, I will share the experience working with different departments and how we achieve better result by leveraging knowledge from every domain.
more
沈聖峰

沈聖峰 生態學的問題意識與資料分析

以巨觀生態學資料分析為例,探討由資料分析到理論模式建立,發展新的巨觀生態學理論的過程。
more
紀懷新

紀懷新 Optimizing for User Experience with Data Science and Machine Learning

Understanding users and optimizing for user experience are critical parts of building successful apps and services. While there had been a tremendous amount of past work studying user and social interactions, in practice, it wasn’t until quite recently that researchers are able to study these interaction mechanisms at scale easily. In this talk, I will illustrate data-driven approaches to understand what are happy engaged users, and present case studies of how we utilize novel machine learning techniques to optimize for long-term user engagements in practice.
more
高宏宇

高宏宇 Deeper Text Mining

本演講將針對深度學習在文字探勘問題上的應用。從命名單元的辨識與擷取中條件亂數場域(conditional random field, CRF)的應用,對話生成模型的差異,到網路謠言偵測三個議題來討論機器學習與深度學習在文字探勘領域的發揮。
more
洪士灝

洪士灝 為 AI 打造系統、用 AI 設計系統

人工智慧(AI)蔚為熱潮,應用面不斷地擴張,然而AI要真正落實於產業,還是必須仰賴優質的系統軟硬體統合設計,以及高度優化的系統晶片,才能孕育出有高度競爭力,乃至於造就破壞性創新的AI產品與服務。如今用GPU加速機器學習,已是兵家常事,Google早在2014開始研發高效低耗能的Tensorflow處理器,如今更是百家爭鳴,不僅為AI設計系統,更以AI輔助設計系統。然而這些並非傳統的硬體設計,要能夠集結各種人才,針對應用的特性,整合與創新軟硬體,才能引領潮流。
more
洪瑞鴻

洪瑞鴻 淺談深度計算基因體學

Thanks to the advance of many ground breaking technologies in sequencing, the scope of Genomics has expanded several times in recent years. These extremely high throughput technologies generate information in an unprecedented rate; however, leveraging these data to facilitate the understanding to the messages encoded in DNA is still challenging. Although with a complete three-billion-base human genome sequence in hand, we human embarrassingly know only next to nothing to the basis of Genomics. Just before all of us are submerged by the tsunami of data, deep learning, the once glimmering machine learning discipline, has resurrected and bailed us out. With the help of GPU computing and Data Science, we can now let machines discover the mechanisms underlying biological phenomena and pathways with only little domain knowledge. Welcome to the era of Deep Computational Genomics!
more
洪智傑

洪智傑 智慧城市的前瞻基礎建設:軌跡資料分析系統

為了建造一個智慧城市,資料分析系統可以稱為是其中的前瞻基礎建設,是不可或缺的一環。其中,軌跡資料是其中來源最多也最具有廣泛應用價值的資料。由於軌跡資料包含使用者位置的資料。這些軌跡資料可以反映使用者如何在城市中活動、如何利用公共運輸設施、如何與城市中的各個角色互動等,因此可以用來讓智慧城市的各項服務品質更好、更有智慧。在此演講中,我們將分享關於軌跡資料系統處理的研究及經驗。其中包含了大規模的即時軌跡資料建模、批次軌跡資料簡化技巧、以及軌跡資料豐富化及應用等,以及針對在GPS軌跡以及捷運卡的資料集上的應用進行案例探討。此演講也希望聽眾在此演講之後能夠對軌跡資料分析於智慧城市的應用有更深一層認識。
more
馬匡六

馬匡六 Big Data Visualization

Advanced computing and imaging technologies enable scientists to study natural phenomena at unprecedented precision, resulting in an explosive growth of data. The size of the collected information about the Internet and mobile device users is expected to be even greater. To make sense and maximize utilization of such vast amounts of data for knowledge discovery and decision making, we need a new set of tools beyond conventional data mining and statistical analysis methods. Visualization transforms large quantities of, often multiple-dimensional, data into graphical representations that exploit the high-bandwidth channel of the human visual system, leveraging the brain's remarkable ability to detect patterns and draw inferences. It has been shown very effective in understanding large, complex data, and thus become an indispensable tool for many areas of research and practice. I will present several use cases of visualization based on new concepts and techniques that my group at UC Davis has introduced to further advance the visualization technology as a powerful discovery and communication tool.
more
孫民

孫民 Training a Deep Agent to See and Interact

An intelligent agent should have the abilities to see and interact with the world in many different ways. In this talk, I summarize our recent work on seeing and interacting using language (e.g., video captioning), interacting by taking actions in specific applications (e.g., viewing angle selection in 360 videos), and interacting by attacking other agents. Ultimately, we long for an embodied intelligent agent to assist us in our daily life. Hence, we also propose a system to anticipate human intention in order to proactively provide assistance.
more
孫玉峰

孫玉峰 Data Science on the Field-資料科學如何影響運動產業?認識賽伯計量學

看別人上太空,我們真的只能殺豬公嗎?當我們看到國際賽事與職業運動場域上越來越多科學化的方式來幫助增強選手以及團隊的表現,不免讓人捫心自問,運動科學在台灣的腳步是不是走得太慢了些?運動科學背後有著龐大的資料科學基礎,這次的演講將會藉由收集運動數據歷史最為悠久的美國職棒大聯盟(MLB)的案例以及分析做為借鏡,嘗試讓台灣的運動產業能夠對於資料科學如何幫助運動科學有著近一步的認識。
more
許之凡

許之凡 當人臉遇到機器學習

凱吉哥真的是凱吉哥?人臉偵測與辨識從古至今一直是電腦視覺領域中非常熱門的題目,而解析人臉的第一步即為找出各部件 (眼睛、鼻子、嘴巴) 的位置,再更進一步解析其外型,利用數個關鍵點來代表各種不同的外型,這些關鍵點無論在哪個臉部部件中一般被統稱為"臉部特徵點" (facial landmarks)。臉部特徵點的應用繁多,從頭部姿勢辨識、虛擬化妝、至臉部外型交換,今年 iPhone X 甚至利用 FaceID 臉部辨識技術來進行安全認證,解除手持式裝置的安全鎖。講者將在本演講中,介紹近幾年如何利用機器學習與深度學習的方式在影像中尋找臉部特徵點的位置,進而介紹如何利用機器學習模型來改變人臉的外型。
more
許懷中

許懷中 魔球之外,談數據分析在棒球比賽臨場之應用

自 1977 年 Bill James開啟了近代棒球統計的濫觴以來,賽柏計量學 (Sabermetrics) 經過數十年的發展,已經廣泛地成為棒球先進國家、頂尖職業聯評估球員表現、構築球隊戰力的理論基礎。 而在臨場戰術的決策、投打對決的電光石火之間,自然也少不了數據統計的參與空間,從直觀上根據打者打擊落點進行的守位調整 (Defensive Shift)、或者對於打者冷熱打擊區域的統計,以及計算使用戰術是否可以獲得利益的數據統計,數據統計確實地影響著棒球場上的一切決策,而要採用數據所透露的情報抑或是相信自己的直覺,則再再地考驗著球隊教練的智慧。 本演講將藉由實際的分析案例,為各位聽眾剖析賽柏計量學是如何在棒球臨場決策上扮演重要的角色,歡迎熱愛國球的棒球痴、棒球狂們前來一同分享。
more
張仁和

張仁和 你知道我在想什麼嗎?內隱連結測量與心理字詞探勘的應用

自陳報告(self-report)的使用是在行為調查中最常見的方式,然而,自陳報告最大的問題也在於,人們是可以調整自己表達的內容的,例如在面對社會期許下做出政治正確的陳述。為了突破自陳量表的限制,心理學嘗試建立內隱(implicit)的測量指標,來更深層地了解人們內在的種種動機和想法,在本次報告中,即會介紹內隱連結測量以及心理字詞探勘,以及這些指標測量在網路和數位資料上的種種應用可能。
more
張佳彥

張佳彥 Machine Learning in Cyber Security

Machine Learning is a powerful tool to detect the unknown cyber threat. We are going to share our experience of using machine learning in Cyber Security.
more
張俊鴻

張俊鴻 國際級的臨床醫療資料研究實戰介紹: 以健康保險資料庫與美國國立醫學圖書館資料庫為例

美國自歐巴馬總統時代提倡精準醫學, 這需要本土種族研究資料與國際治療實證資料結合,張醫師由臨床醫師的角度介紹臨床問題如何用這兩個資料庫分析
more
頁數: 1 2 3 4 5 6 7 - 每頁 20 筆

歡迎在此登錄您的大名及電子郵件地址,日後任何台灣資料科學協會舉辦的相關活動,我們將會以電子郵件通知您。謝謝。