CrowdFlow: Integrating Machine Learning with Mechanical Turk for Speed-Cost-Quality Flexibility
نویسندگان
چکیده
Many problems involving visual perception, language understanding, or other human abilities can now be solved by computers. This allows the tasks to be done faster and more affordably than humans could do. Furthermore, it means they can be done on-demand, enabling computers to monitor national intelligence information streams, search video footage for missing persons, and perform a wide variety of other services of importance to society. The primary disadvantage is that machines are still not as accurate as humans on many tasks.
منابع مشابه
Active Learning for Crowd-Sourced Databases
Crowd-sourcing has become a popular means of acquiring labeled data for many tasks where humans are more accurate than computers, such as image tagging, entity resolution, or sentiment analysis. However, due to the time and cost of human labor, solutions that solely rely on crowd-sourcing are often limited to small datasets (i.e., a few thousand items). This paper proposes algorithms for integr...
متن کاملActive Learning and Crowd-Sourcing for Machine Translation
In recent years, corpus based approaches to machine translation have become predominant, with Statistical Machine Translation (SMT) being the most actively progressing area. Success of these approaches depends on the availability of parallel corpora. In this paper we propose Active Crowd Translation (ACT), a new paradigm where active learning and crowd-sourcing come together to enable automatic...
متن کاملActive Learning with Efficient Feature Weighting Methods for Improving Data Quality and Classification Accuracy
Many machine learning datasets are noisy with a substantial number of mislabeled instances. This noise yields sub-optimal classification performance. In this paper we study a large, low quality annotated dataset, created quickly and cheaply using Amazon Mechanical Turk to crowdsource annotations. We describe computationally cheap feature weighting techniques and a novel non-linear distribution ...
متن کاملSemi-supervised Word Alignment with Mechanical Turk
Word alignment is an important preprocessing step for machine translation. The project aims at incorporating manual alignments from Amazon Mechanical Turk (MTurk) to help improve word alignment quality. As a global crowdsourcing service, MTurk can provide flexible and abundant labor force and therefore reduce the cost of obtaining labels. An easyto-use interface is developed to simplify the lab...
متن کاملActive Learning with Amazon Mechanical Turk
Supervised classification needs large amounts of annotated training data that is expensive to create. Two approaches that reduce the cost of annotation are active learning and crowdsourcing. However, these two approaches have not been combined successfully to date. We evaluate the utility of active learning in crowdsourcing on two tasks, named entity recognition and sentiment detection, and sho...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010