Maximum Entropy Models with Inequality Constraints: A Case Study on Text Categorization
نویسندگان
چکیده
منابع مشابه
Evaluation and Extension of Maximum Entropy Models with Inequality Constraints
A maximum entropy (ME) model is usually estimated so that it conforms to equality constraints on feature expectations. However, the equality constraint is inappropriate for sparse and therefore unreliable features. This study explores an ME model with box-type inequality constraints, where the equality can be violated to reflect this unreliability. We evaluate the inequality ME model using text...
متن کاملSummary of Text Categorization based on Maximum Entropy Model
Since 1990s, the maximum entropy model has been used in text categorization and achieves good results in Natural Language Processing since its framework and algorithm were established. On the basis of the Maximum Entropy Model, scholars improve it and make a more in-depth study. Using Maximum Entropy Model for text sentiment categorization has become a hot research topic in recent years. In thi...
متن کاملMaximum entropy estimation in economic models with linear inequality restrictions
In this paper, we use maximum entropy to estimate the parameters in an economic model. We demonstrate the use of the generalized maximum entropy (GME) estimator, describe how to specify the GME parameter support matrix, and examine the sensitivity of GME estimates to the parameter and error bounds. We impose binding inequality restrictions through the GME parameter support matrix and develop a ...
متن کاملAutomatic Text Categorization: Case Study
Text Categorization is a process of classifying documents with regard to a group of one or more existent categories [1] according to themes or concepts present in their contents. The most common application of it is in Information Retrieval Systems (IRS) to document indexing [2]. The organization of text in categories allow the user to limit the target of a search submitted to IRS, to explore t...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Machine Learning
سال: 2005
ISSN: 0885-6125,1573-0565
DOI: 10.1007/s10994-005-0911-3