Active Learning of Classification Models with Likert-Scale Feedback

نویسندگان

  • Yanbing Xue
  • Milos Hauskrecht
چکیده

Annotation of classification data by humans can be a time-consuming and tedious process. Finding ways of reducing the annotation effort is critical for building the classification models in practice and for applying them to a variety of classification tasks. In this paper, we develop a new active learning framework that combines two strategies to reduce the annotation effort. First, it relies on label uncertainty information obtained from the human in terms of the Likert-scale feedback. Second, it uses active learning to annotate examples with the greatest expected change. We propose a Bayesian approach to calculate the expectation and an incremental SVM solver to reduce the time complexity of the solvers. We show the combination of our active learning strategy and the Likert-scale feedback can learn classification models more rapidly and with a smaller number of labeled instances than methods that rely on either Likert-scale labels or active learning alone.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Mohs Surgical Reconstruction Educational Activity: a resident education tool

BACKGROUND Surgical reconstructive planning following Mohs surgery can be a difficult subject for dermatology residents to master. Prior research demonstrates that active learning is preferred and more effective compared to passive learning models and that dermatology residents desire greater complexity and volume in surgical training. We present a novel, active, problem-based learning tool for...

متن کامل

کمیت و کیفیت بازخورد در محیط بالینی از نظر دانشجویان مامایی

Introduction: Although feedback is recognized as an integral part of the educational process; it has challenged the educational network complexity from one learning community to another,. Providing training programs to empower teachers and learners in this respect requires identifying educational gaps in the provision of effective feedback. Accordingly, this study is designed and implemented to...

متن کامل

The hybrid model of clinical skills teaching and the learning theories behind it.

Introduction: There are two popular methods of clinical skillsteaching. One is Peyton’s method, and the other one is RobertGagne’s method. A hybrid model which is a combination ofboth teaching methods is developed and implemented at AvalonUniversity School of Medicine in Clinical Skills. The aim of thestudy was to evaluate the hybrid model of clinical skills teachi...

متن کامل

نقش بازخورد در آموزش بالینی: اصول، راهبردها و مدل‌ها

Introduction: Feedback is an inseparable and integral part of learning. It promotes students' learning towards achieving goals. In the training of health professions, little feedback or a paucity of feedback can be observed. Giving correct feedback is another important matter that must be taken into account. Therefore, the aim of this study was to present the principles of feedback, its effecti...

متن کامل

Life-wise Language Learning Textbooks: Construction and Validation of an Emotional Abilities Scale through Rasch Modeling

Underlying the recently developed notions of applied ELT and life syllabus is the idea that language classes should give precedence to learners’ life qualities, for instance emotional intelligence (EI), over and above their language skills. By so doing, ELT is ascribed an autonomous status and ELT classes can lavish their full potentials to the learners. With that in mind, this study aimed to d...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • Proceedings of the ... SIAM International Conference on Data Mining. SIAM International Conference on Data Mining

دوره 2017  شماره 

صفحات  -

تاریخ انتشار 2017