Notes on Phrasal Indexing: JSCB Evaluation Experiments at NTCIR AD HOC

نویسنده

  • Sumio Fujita
چکیده

The evaluation experiments of the JSCB team are described with a focus on noun phrase indexing and its weighting issues in ad hoc text retrieval. Experiments on the effects of supplemental noun phrase indexing in view of the effect of various length of queries are reported. The results show that the noun phrase indexing outperforms single word only indexing with long queries while single word only indexing performs slightly better with short queries. A new weighting method for phrasal terms is also evaluated and improvement is observed.

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

Notes on the Limits of CLIR Effectiveness: NTCIR-2 Evaluation Experiments at Justsystem

NTCIR-2 evaluation experiments at the Justsystem site are described with a focus on comparative study of CLIR effectiveness with monolingual retrieval effectiveness of the same retrieval engine. Experiments on the effects of phrasal translation, indexing of translated phrasal terms, pre-translation feedback and parallel documents feedback in diverse retrieval settings, are reported. The results...

متن کامل

NCU in Bilingual Information Retrieval Experiments at NTCIR-6

In this paper, we present the mono-lingual and bilingual ad-hoc information retrieval experimental results at NTCIR-6. This year we compare two different word tokenization levels for indexing, namely, unigram, and overlapping bigram. The two famous information retrieval models, i.e., language model, and BM-25 were adopted in our study. In the mono-lingual results show that our method achieved t...

متن کامل

R2D2 at NTCIR 2 Ad-hoc Task: Relevance-based Superimposition Model for IR

This paper describes our evaluation experiments for NTCIR 2 ad-hoc task. We developed a retrieval system using the Relevance-based Superimposition (RS) model, in which document vectors are modified based on the relevance of the documents. The major focus of this year is on combination of the RS model and query expansion (QE). We submitted fully automatic ad-hoc results brought by different para...

متن کامل

Sampling Precision to Depth 9000: Evaluation Experiments at NTCIR-6

We describe evaluation experiments conducted by submitting retrieval runs for the Chinese, Japanese and Korean Single Language Information Retrieval subtasks of the Cross-Lingual Information Retrieval (CLIR) Task of the 6th NII Test Collection for IR Systems Workshop (NTCIR-6). We show that a Generalized Success@10 measure exposes a downside of the blind feedback technique that is overlooked by...

متن کامل

Analysis of the Usage of Japanese Segmented Texts in NTCIR Workshop 2

In this paper, we report on the usage of Japanese segmented texts and analyze the submitted search results to NTCIR Workshop 2, which used these texts. In these texts, each sentence is segmented into terms and term components (similar to phrases and words). However, the sizes of terms are inconsistent in the texts; e.g., some terms that should be decomposed into term components remain as terms....

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 1999