• ETRI Journal
• /
• v.38 no.3
• /
• pp.487-493
• /
• 2016

Two new methods are proposed for an unsupervised adaptation of a language model (LM) with a single sentence for automatic transcription tasks. At the training phase, training documents are clustered by a method known as Latent Dirichlet allocation (LDA), and then a domain-specific LM is trained for each cluster. At the test phase, an adapted LM is presented as a linear mixture of the now trained domain-specific LMs. Unlike previous adaptation methods, the proposed methods fully utilize a trained LDA model for the estimation of weight values, which are then to be assigned to the now trained domain-specific LMs; therefore, the clustering and weight-estimation algorithms of the trained LDA model are reliable. For the continuous speech recognition benchmark tests, the proposed methods outperform other unsupervised LM adaptation methods based on latent semantic analysis, non-negative matrix factorization, and LDA with n-gram counting.

• Korean Journal of Cognitive Science
• /
• v.26 no.4
• /
• pp.377-392
• /
• 2015

Korean semantic role labeling (SRL) is usually performed by a machine learning and requires a lot of corpus. However, the Korean PropBank used in Korean SRL system is less than PropBank. It leads to a low performance. Therefore, we expand the annotated corpus and verb frames for Korean SRL system to expand the Korean PropBank corpus. Most of the SRL system have a domain-dependent performance so, the performance may decrease if domain was changed. In this paper, we use the domain adaptation technique to reduce decreasing performance with the existing corpus and the small size of new domain corpus. We apply the domain adaptation technique to Structural SVM and Deep Neural Network. The experimental result show the effectiveness of the domain adaptation technique.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1208-1215
• /
• 2017

Recently, people's attention and worries about fine particulate matter have been increasing. Due to the construction and maintenance costs, there are insufficient air quality monitoring stations. As a result, people have limited information about the concentration of fine particulate matter, depending on the location. Studies have been undertaken to estimate the fine particle concentrations in areas without a measurement station. Yet there are limitations in that the estimate cannot take account of other factors that affect the concentration of fine particle. In order to solve these problems, we propose a framework for estimating the concentration of fine particulate matter of a specific area using meteorological data and traffic data. Since there are more grids without a monitor station than grids with a monitor station, we used a domain adversarial neural network based on the domain adaptation method. The features extracted from meteorological data and traffic data are learned in the network, and the air quality index of the corresponding area is then predicted by the generated model. Experimental results demonstrate that the proposed method performs better as the number of source data increases than the method using conditional random fields.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.2
• /
• pp.73-81
• /
• 2024

Recently, mobile manipulators have been utilized in agriculture industry for weed removal and harvest automation. This paper proposes a domain adaptive fruit detection method for harvest automation, by utilizing OWL-ViT model which is an open-vocabulary object detection model. The vision-language model can detect objects based on text prompt, and therefore, it can be extended to detect objects of undefined categories. In the development of deep learning models for real-world problems, constructing a large-scale labeled dataset is a time-consuming task and heavily relies on human effort. To reduce the labor-intensive workload, we utilized a large-scale public dataset as a source domain data and employed a domain adaptation method. Adversarial learning was conducted between a domain discriminator and feature extractor to reduce the gap between the distribution of feature vectors from the source domain and our target domain data. We collected a target domain dataset in a real-like environment and conducted experiments to demonstrate the effectiveness of the proposed method. In experiments, the domain adaptation method improved the AP50 metric from 38.88% to 78.59% for detecting objects within the range of 2m, and we achieved 81.7% of manipulation success rate.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2265-2267
• /
• 1998

This paper deals with study of Fast Speaker Adaptation Type Speech Recognition, and to analyze speech signal efficiently in time domain and time-frequency domain, utilizes SCONN[1] with Speech Signal Process suffices for Fast Speaker Adaptation Type Speech Recognition, and examined Speech Recognition to investigate adaptation of system, which has speech data input after speaker dependent recognition test.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.5
• /
• pp.2590-2606
• /
• 2017

Generally, research of classical image classification algorithms assume that training data and testing data are derived from the same domain with the same distribution. Unfortunately, in practical applications, this assumption is rarely met. Aiming at the problem, a domain adaption image classification approach based on multi-sparse representation is proposed in this paper. The existences of intermediate domains are hypothesized between the source and target domains. And each intermediate subspace is modeled through online dictionary learning with target data updating. On the one hand, the reconstruction error of the target data is guaranteed, on the other, the transition from the source domain to the target domain is as smooth as possible. An augmented feature representation produced by invariant sparse codes across the source, intermediate and target domain dictionaries is employed for across domain recognition. Experimental results verify the effectiveness of the proposed algorithm.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.11 no.2
• /
• pp.234-242
• /
• 1987

The solution of shape design problems based on variational analysis has been approached by using the domain adaptive method. The objective of the structural shape design is to minimize the weight within a bound on local stress measure, or to minimize the maximum local stress measure within a bound on the weight. A derived optimality condition in both design problems requires that the unit mutual energy has constant value along the design boundary. However, the condition for constant stress on the design boundary was used in computation since the computed mutual energy oscillates severely on the boundary. A two step iteration scheme using domain adaptation was presented as a computational method to slove the example designs of elastic structures. It was also shown that remeshing by grid adaptation was effective to reduce oscillatory behavior on the design boundary.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.6C
• /
• pp.423-433
• /
• 2005

We present power(in frequency domain) and rate adaptation(in time domain) schemes in multicarrier (MC) direct-sequence code-division multiple-access(DS/CDMA) communications. Utilizing channel state information from the receiver, the adaptation schemes allocate power the user's sub-band with the largest channel gain. In the time domain, the transmission data rate is adapted for a desired transmission quality. In the case of single-user channels, a closed-form expression is derived for an optimal time domain power adaptation that minimizes the average bit error rate(BER). Channel inversion power adaptation is found to provide nearly optimal performance in this case, as the number of sub-bands or available average transmission power increase. Analysis and simulation results show the BER performance of the proposed power and rate adaptations with fixed average transmission power significantly improves the performance over the power allocation in the frequency domain only. Also, we compare the performance of the proposed power and rate adaptation schemes in MC-DS/CDMA systems with that of power and rate adapted single carrier DS/CDMA systems with RAKE receiver.

• Journal of KIISE
• /
• v.42 no.4
• /
• pp.475-482
• /
• 2015

Developing a high-performance Semantic Role Labeling (SRL) system for a domain requires manually annotated training data of large size in the same domain. However, such SRL training data of sufficient size is available only for a few domains. Performances of Korean SRL are degraded by almost 15% or more, when it is directly applied to another domain with relatively small training data. This paper proposes two techniques to minimize performance degradation in the domain transfer. First, a domain adaptation algorithm for Korean SRL is proposed which is based on the prior model that is one of domain adaptation paradigms. Secondly, we proposed to use simplified features related to morphological and syntactic tags, when using small-sized target domain data to suppress the problem of data sparseness. Other domain adaptation techniques were experimentally compared to our techniques in this paper, where news and Wikipedia were used as the sources and target domains, respectively. It was observed that the highest performance is achieved when our two techniques were applied together. In our system's performance, F1 score of 64.3% was considered to be 2.4~3.1% higher than the methods from other research.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2E
• /
• pp.51-55
• /
• 2004

In this paper, we propose a style-specific language model adaptation scheme using n-gram based tf*idf similarity for Korean spontaneous speech recognition. Korean spontaneous speech shows especially different style-specific characteristics such as filled pauses, word omission, and contraction, which are related to function words and depend on preceding or following words. To reflect these style-specific characteristics and overcome insufficient data for training language model, we estimate in-domain dependent n-gram model by relevance weighting of out-of-domain text data according to their n-. gram based tf*idf similarity, in which in-domain language model include disfluency model. Recognition results show that n-gram based tf*idf similarity weighting effectively reflects style difference.