Learning Dependencies between Case Frame Slots

نویسندگان

  • Hang Li
  • Naoki Abe
چکیده

We address the problem of automatically acquiring case frame patterns (selectional patterns) from large corpus data. In particular, we propose a method of learning dependencies between case frame slots. We view the problem of learning case frame patterns as that of learning multi-dimensional discrete joint distributions, where random variables represent case slots. We then formalize the dependencies between case slots as the probabilistic dependencies between these random variables. Since the number of parameters in a multi-dimensional joint distribution is exponential, it is infeasible to accurately estimate them in practice. To overcome this difficulty, we settle with approximating the target joint distribution by the product of low order component distributions, based on corpus data. In particular we propose to employ an efficient learning algorithm based on the MDL principle to realize this task. Our experimental results indicate that for certain classes of verbs, the accuracy achieved in a disambiguation experiment is improved by using the acquired knowledge of dependencies.

برای دانلود رایگان 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

Bayesian Network Models of Subcategorization and Their MDL-Based Learning from Corpus

We formalize two probabilistic models of verbal subcategorization based on the Bayesian network which treat dependencies among case slots as well as the class generalization of adjunct/argument nouns. We implement algorithms for obtaining locally optimal models based on the MDL principle and evaluate the obtained models in terms of syntactic disambiguation task. In the task, we compare the two ...

متن کامل

Improve Downlink Burst Allocation to Achieve High Frame Utilization of Mobile WiMAX (802.16e)

The burst allocation algorithm is responsible about calculating the appropriate dimensions and location of each user’s data to construct the bursts in the downlink subframe in term of the number of slots for each user. Burst allocation is one of the performance parameter which influences the mobile WiMAX systems, due to resource wastage in the form of unused and unallocated slots per frame whic...

متن کامل

AFAST: An Automatic Frames Acquisition System

This paper describes an unsupervised strategy to acquire lexico-semantic frames (LSFs) of verbs from sentential parsed corpora (in syntactic level). The problems of acquiring LSFs consist of verb senses ambiguity, diversity of linguistic usages, and lack of completed frame slots in a single sentence. We propose an specific clustering technique based on the Minimum Description Length (MDL) princ...

متن کامل

Broadcast scheduling in packet radio networks using mixed tabu-greedy algorithm - Electronics Letters

Introduction: The broadcast scheduling problem (BSP) is to find a collision-free scheduling of transmissions of all the stations in a minimum number of time slots. The final arrangement of the station transmissions into their allocated time slots is called a frame. Consider a packet radio network (PRN) represented by a graph G1⁄4 (V, E), where V is the set of nodes and E the set of edges. An N ...

متن کامل

Learning semantic hierarchy with distributed representations for unsupervised spoken language understanding

We study the problem of unsupervised ontology learning for semantic understanding in spoken dialogue systems, in particular, learning the hierarchical semantic structure from the data. Given unlabelled conversations, we augment a frame-semantic based unsupervised slot induction approach with hierarchical agglomerative clustering to merge topically-related slots (e.g., both slots “direction” and...

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

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1996