Enhancing the Input Representation: From Complexity to Simplicity
نویسندگان
چکیده
We introduce an efficient algorithm for mining informative combinations of attribute-values for a given task. We use informative attributevalues to enhance the input representation of data. We apply our approach to coreference resolution using a simple set of attributes like syntactic roles and string match. With the enhanced representation, a simple coreference model outperforms more complex state-ofthe-art models by a large margin. The use of the enhanced representation results in robust improvements in both in-domain and out-ofdomain evaluations.
منابع مشابه
Changing the Role of Teacher according to Complexity Theory: From Representation to Facilitating Emergence
The present study seeks to rethink the role of the teacher in the teaching-learning process according to the complexity theory. First, the role of the teacher is explained in the traditional vision of Comenius and Dewey's critical insight and then the role of the teacher is discussed in the complexity theory. Then, the teacher’s image as an emergence facilitator is suggested instead of their im...
متن کاملThe Effect of Comprehensible Input and Comprehensible Output on the Accuracy and Complexity of Iranian EFL Learners’ Oral Speech
This study aimed at investigating the relative impact of comprehensible input and comprehensible output on the development of grammatical accuracy and syntactic complexity of Iranian EFL learners’ oral production. Participants were 60 female EFL learners selected from a whole population pool of 80 based on the standard test of IELTS. To investigate the research questions, the participants were ...
متن کاملInput-induced Variation in EFL Learners’ Oral Production in Terms of Complexity, Accuracy, and Fluency
Researchers have extensively studied phenomena that affect a second language learner’s oral production while there is scant evidence about input-related factors. Accordingly, the present study sought to investigate how variation in oral production is caused by the input they receive from different course materials. To this end, the study included a micro-evaluation study of three course materia...
متن کاملImage Classification via Sparse Representation and Subspace Alignment
Image representation is a crucial problem in image processing where there exist many low-level representations of image, i.e., SIFT, HOG and so on. But there is a missing link across low-level and high-level semantic representations. In fact, traditional machine learning approaches, e.g., non-negative matrix factorization, sparse representation and principle component analysis are employed to d...
متن کاملEnhancing Efficiency of Neural Network Model in Prediction of Firms Financial Crisis Using Input Space Dimension Reduction Techniques
The main focus in this study is on data pre-processing, reduction in number of inputs or input space size reduction the purpose of which is the justified generalization of data set in smaller dimensions without losing the most significant data. In case the input space is large, the most important input variables can be identified from which insignificant variables are eliminated, or a variable ...
متن کاملذخیره 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 ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1708.00160 شماره
صفحات -
تاریخ انتشار 2017