Unsupervised Word Alignment Using Frequency Constraint in Posterior Regularized EM

2014 ◽  
Author(s):  
Hidetaka Kamigaito ◽  
Taro Watanabe ◽  
Hiroya Takamura ◽  
Manabu Okumura
Keyword(s):  
Word Alignment ◽  
Frequency Constraint ◽  
Regularized Em

scholarly journals Unsupervised Word Alignment Using Frequency Constraint in Posterior Regularized EM

2016 ◽  
Vol 23 (4) ◽  
pp. 327-351
Author(s):  
Hidetaka Kamigaito ◽  
Taro Watanabe ◽  
Hiroya Takamura ◽  
Manabu Okumura ◽  
Eiichiro Sumita
Keyword(s):  
Word Alignment ◽  
Frequency Constraint ◽  
Regularized Em

Latent Domain Word Alignment for Heterogeneous Corpora

2015 ◽  
Author(s):  
Hoang Cuong ◽  
Khalil Sima'an
Keyword(s):  
Word Alignment

A Constrained Viterbi Relaxation for Bidirectional Word Alignment

2014 ◽  
Author(s):  
Yin-Wen Chang ◽  
Alexander M. Rush ◽  
John DeNero ◽  
Michael Collins
Keyword(s):  
Word Alignment

scholarly journals Design and Analysis of Super Wideband Antenna for Microwave Applications

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 477
Author(s):  
Warsha Balani ◽  
Mrinal Sarvagya ◽  
Ajit Samasgikar ◽  
Tanweer Ali ◽  
Pradeep Kumar
Keyword(s):  
Communication Systems ◽  
Equivalent Circuit Model ◽  
Wireless Communication Systems ◽  
Impedance Bandwidth ◽  
Frequency Range ◽  
Fractional Bandwidth ◽  
Wide Bandwidth ◽  
Wireless Applications ◽  
Frequency Constraint ◽  
Time Domain Characterization

In this article, a compact concentric structured monopole patch antenna for super wideband (SWB) application is proposed and investigated. The essential characteristics of the designed antenna are: (i) to attain super-wide bandwidth characteristics, the proposed antenna is emerged from a traditional circular monopole antenna and has obtained an impedance bandwidth of 38.9:1 (ii) another important characteristic of the presented antenna is its larger bandwidth dimension ratio (BDR) value of 6596 that is accomplished by augmenting the electrical length of the patch. The electrical dimension of the proposed antenna is 0.18λ×0.16λ (λ corresponds to the lower end operating frequency). The designed antenna achieves a frequency range from 1.22 to 47.5 GHz with a fractional bandwidth of 190% and exhibiting S11 < −10 dB in simulation. For validating the simulated outcomes, the antenna model is fabricated and measured. Good conformity is established between measured and simulated results. Measured frequency ranges from 1.25 to 40 GHz with a fractional bandwidth of 188%, BDR of 6523 and S11 < −10 dB. Even though the presented antenna operates properly over the frequency range from 1.22 to 47.5 GHz, the results of the experiment are measured till 40 GHz because of the high-frequency constraint of the existing Vector Network Analyzer (VNA). The designed SWB antenna has the benefit of good gain, concise dimension, and wide bandwidth above the formerly reported antenna structures. Simulated gain varies from 0.5 to 10.3 dBi and measured gain varies from 0.2 to 9.7 dBi. Frequency domain, as well as time-domain characterization, has been realized to guide the relevance of the proposed antenna in SWB wireless applications. Furthermore, an equivalent circuit model of the proposed antenna is developed, and the response of the circuit is obtained. The presented antenna can be employed in L, S, C, X, Ka, K, Ku, and Q band wireless communication systems.

scholarly journals Phrase Table Combination Based on Symmetrization of Word Alignment for Low-Resource Languages

2021 ◽  
Vol 11 (4) ◽  
pp. 1868
Author(s):  
Sari Dewi Budiwati ◽  
Al Hafiz Akbar Maulana Siagian ◽  
Tirana Noor Fatyanosa ◽  
Masayoshi Aritsugi
Keyword(s):  
Language Model ◽  
Word Alignment ◽  
Low Resource ◽  
Relative Improvement ◽  
Direct Translation ◽  
The Common ◽  
Target Source

Phrase table combination in pivot approaches can be an effective method to deal with low-resource language pairs. The common practice to generate phrase tables in pivot approaches is to use standard symmetrization, i.e., grow-diag-final-and. Although some researchers found that the use of non-standard symmetrization could improve bilingual evaluation understudy (BLEU) scores, the use of non-standard symmetrization has not been commonly employed in pivot approaches. In this study, we propose a strategy that uses the non-standard symmetrization of word alignment in phrase table combination. The appropriate symmetrization is selected based on the highest BLEU scores in each direct translation of source–target, source–pivot, and pivot–target of Kazakh–English (Kk–En) and Japanese–Indonesian (Ja–Id). Our experiments show that our proposed strategy outperforms the direct translation in Kk–En with absolute improvements of 0.35 (a 11.3% relative improvement) and 0.22 (a 6.4% relative improvement) BLEU points for 3-gram and 5-gram, respectively. The proposed strategy shows an absolute gain of up to 0.11 (a 0.9% relative improvement) BLEU points compared to direct translation for 3-gram in Ja–Id. Our proposed strategy using a small phrase table obtains better BLEU scores than a strategy using a large phrase table. The size of the target monolingual and feature function weight of the language model (LM) could reduce perplexity scores.

A Low-Area and Low-Power Comma Detection and Word Alignment Circuits for JESD204B/C Controller

2021 ◽  
pp. 1-11
Author(s):  
Peng Yin ◽  
Zhou Shu ◽  
Yingjun Xia ◽  
Tianmei Shen ◽  
Xiao Guan ◽  
...  
Keyword(s):  
Low Power ◽  
Word Alignment ◽  
Low Area

Does GIZA++ Make Search Errors?

2010 ◽  
Vol 36 (3) ◽  
pp. 295-302 ◽  
Author(s):  
Sujith Ravi ◽  
Kevin Knight
Keyword(s):  
Machine Translation ◽  
Statistical Machine Translation ◽  
Word Alignment ◽  
Alignment Algorithm

Word alignment is a critical procedure within statistical machine translation (SMT). Brown et al. (1993) have provided the most popular word alignment algorithm to date, one that has been implemented in the GIZA (Al-Onaizan et al., 1999) and GIZA++ (Och and Ney 2003) software and adopted by nearly every SMT project. In this article, we investigate whether this algorithm makes search errors when it computes Viterbi alignments, that is, whether it returns alignments that are sub-optimal according to a trained model.

Sign in / Sign up

Export Citation Format

Share Document