Does Size Matter - How Much Data is Required to Train a REG Algorithm?
نویسندگان
چکیده
In this paper we investigate how much data is required to train an algorithm for attribute selection, a subtask of Referring Expressions Generation (REG). To enable comparison between different-sized training sets, a systematic training method was developed. The results show that depending on the complexity of the domain, training on 10 to 20 items may already lead to a good performance.
منابع مشابه
روشی جدید برای عضویتدهی به دادهها و شناسایی نوفه و دادههای پرت با استفاده از ماشین بردار پشتیبان فازی
Support Vector Machine (SVM) is one of the important classification techniques, has been recently attracted by many of the researchers. However, there are some limitations for this approach. Determining the hyperplane that distinguishes classes with the maximum margin and calculating the position of each point (train data) in SVM linear classifier can be interpreted as computing a data membersh...
متن کاملTrain Scheduling Problem with Consideration of Praying Constraint as an Application of Job Shop Scheduling Problem
The present paper extends the idea of job shop scheduling problem with resting constraints to the train scheduling problem with the Muslim praying considerations. For this purpose, after proposing the new mathematical model, a heuristic algorithm based on the Electromagnetism-Like algorithm (EM) which is well adjusted to scheduling problems is employed to solve the large-size practical cases. T...
متن کاملLearning Preferences for Referring Expression Generation: Effects of Domain, Language and Algorithm
One important subtask of Referring Expression Generation (REG) algorithms is to select the attributes in a definite description for a given object. In this paper, we study how much training data is required for algorithms to do this properly. We compare two REG algorithms in terms of their performance: the classic Incremental Algorithm and the more recent Graph algorithm. Both rely on a notion ...
متن کاملLearning Preferences for Referring Expression Generation: Effects of Domain, Language and Algorithm
One important subtask of Referring Expression Generation (REG) algorithms is to select the attributes in a definite description for a given object. In this paper, we study how much training data is required for algorithms to do this properly. We compare two REG algorithms in terms of their performance: the classic Incremental Algorithm and the more recent Graph algorithm. Both rely on a notion ...
متن کاملFUZZY TRAIN ENERGY CONSUMPTION MINIMIZATION MODEL AND ALGORITHM
Train energy saving problem investigates how to control train's velocity such that the quantity of energy consumption is minimized and some system constraints are satis ed. On the assumption that the train's weights on different links are estimated by fuzzy variables when making the train scheduling strategy, we study the fuzzy train energy saving problem. First, we propose a fuzzy energy ...
متن کامل