• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.57-67
• /
• 2021

Recently, multi-modal deep learning techniques that combine heterogeneous data for deep learning analysis have been utilized a lot. In particular, studies on the synthesis of Text to Image that automatically generate images from text are being actively conducted. Deep learning for image synthesis requires a vast amount of data consisting of pairs of images and text describing the image. Therefore, various data augmentation techniques have been devised to generate a large amount of data from small data. A number of text augmentation techniques based on synonym replacement have been proposed so far. However, these techniques have a common limitation in that there is a possibility of generating a incorrect text from the content of an image when replacing the synonym for a noun word. In this study, we propose a text augmentation method to replace words using word hierarchy information for noun words. Additionally, we performed experiments using MSCOCO data in order to evaluate the performance of the proposed methodology.

• Communications for Statistical Applications and Methods
• /
• v.30 no.3
• /
• pp.343-354
• /
• 2023

Recently, there have been many improvements in general language models using architectures such as GPT-3 proposed by Brown et al. (2020). Nevertheless, training complex models can hardly be done if the number of data is very small. Data augmentation that addressed this problem was more than normal success in image data. Image augmentation technology significantly improves model performance without any additional data or architectural changes (Perez and Wang, 2017). However, applying this technique to textual data has many challenges because the noise to be added is veiled. Thus, we have developed a novel method for performing data augmentation on text data. We divide the data into signals with positive or negative meaning and noise without them, and then perform data augmentation using k-doc augmentation to randomly combine signals and noises from all data to generate new data.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.455-460
• /
• 2023

Text augmentation is a methodology that creates new augmented texts by transforming or generating original texts for the purpose of improving the performance of NLP models. However existing text augmentation techniques have limitations such as lack of expressive diversity semantic distortion and limited number of augmented texts. Recently text augmentation using large language models and few-shot learning can overcome these limitations but there is also a risk of noise generation due to incorrect generation. In this paper, we propose a text augmentation method called TAGS that generates multiple candidate texts and selects the appropriate text as the augmented text. TAGS generates various expressions using few-shot learning while effectively selecting suitable data even with a small amount of original text by using contrastive learning and similarity comparison. We applied this method to task-oriented chatbot data and achieved more than sixty times quantitative improvement. We also analyzed the generated texts to confirm that they produced semantically and expressively diverse texts compared to the original texts. Moreover, we trained and evaluated a classification model using the augmented texts and showed that it improved the performance by more than 0.1915, confirming that it helps to improve the actual model performance.

• ETRI Journal
• /
• v.42 no.5
• /
• pp.761-772
• /
• 2020

This paper presents an automatic proficiency assessment method for a non-native Korean read utterance using bidirectional long short-term memory (BLSTM)-based acoustic models (AMs) and speech data augmentation techniques. Specifically, the proposed method considers two scenarios, with and without prompted text. The proposed method with the prompted text performs (a) a speech feature extraction step, (b) a forced-alignment step using a native AM and non-native AM, and (c) a linear regression-based proficiency scoring step for the five proficiency scores. Meanwhile, the proposed method without the prompted text additionally performs Korean speech recognition and a subword un-segmentation for the missing text. The experimental results indicate that the proposed method with prompted text improves the performance for all scores when compared to a method employing conventional AMs. In addition, the proposed method without the prompted text has a fluency score performance comparable to that of the method with prompted text.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.3-4
• /
• 2023

자연어처리는 수많은 정보가 수집된 전자의무기록의 비정형 데이터에서 유의미한 정보나 패턴 등을 추출해 의료진의 의사결정을 지원하고, 환자에게 더 나은 진단이나 치료 등을 지원할 수 있어 큰 잠재력을 가지고 있다. 그러나 전자의무기록은 개인정보와 같은 민감한 정보가 다수 포함되어 있어 접근하기 어렵고, 이로 인해 충분한 양의 데이터를 확보하기 어렵다. 따라서 본 논문에서는 신뢰할 수 있는 의료 합성 텍스트를 생성하기 위해 ChatGPT 기반 의료 텍스트 증강 도구를 개발하였다. 이는 사용자가 입력한 실제 의료 텍스트로 의료 합성 데이터를 생성한다. 이를 위해, 적합한 프롬프트와 의료 텍스트에 대한 전처리 방법을 탐색하였다. ChatGPT 기반 의료 텍스트 증강 도구는 입력 텍스트의 핵심 키워드를 잘 유지하였고, 사실에 기반한 의료 합성 텍스트를 생성할 수 있다는 것을 확인할 수 있었다.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.6
• /
• pp.257-264
• /
• 2019

Recently, deep neural network based approaches have shown a good performance for various fields of natural language processing. A huge amount of training data is essential for building a deep neural network model. However, collecting a large size of training data is a costly and time-consuming job. A data augmentation is one of the solutions to this problem. The data augmentation of text data is more difficult than that of image data because texts consist of tokens with discrete values. Generative adversarial networks (GANs) are widely used for image generation. In this work, we generate sentimental texts by using one of the GANs, CS-GAN model that has a discriminator as well as a classifier. We evaluate the usefulness of generated sentimental texts according to various measurements. CS-GAN model not only can generate texts with more diversity but also can improve the performance of its classifier.

• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1214-1222
• /
• 2020

Pervasive augmented reality (AR) technology can be used to efficiently search for the required information regarding products in stores through text augmentation in an Internet of Things (IoT) environment. The evolution of context awareness and image processing technologies are the main driving forces that realize this type of AR service. One of the problems to be addressed in the service is that augmented objects are fixed and cannot be replaced efficiently in real time. To address this problem, a real-time mobile AR framework is proposed. In this framework, an optimal object to be augmented is selected based on object similarity comparison, and the augmented objects are efficiently managed using distributed metadata servers to adapt to the user requirements, in a given situation. To evaluate the feasibility of the proposed framework, a prototype system was implemented, and a qualitative evaluation based on questionnaires was conducted. The experimental results show that the proposed framework provides a better user experience than existing features in smartphones, and through fast AR service, the users are able to conveniently obtain additional information on products or objects.

• Korean Journal of Computational Design and Engineering
• /
• v.12 no.3
• /
• pp.171-181
• /
• 2007

Ubiquitous computing is a vision of our future computing lifestyle in which computer systems seamlessly integrate into our everyday lives, providing services and information in anywhere and anytime fashion. Augmented reality (AR) can naturally complement ubiquitous computing by providing an intuitive and collaborative visualization and simulation interface to a three-dimensional information space embedded within physical reality. This paper presents a service framework and its applications for providing context-aware u-car maintenance services using augmented reality, which can support a rich set of ubiquitous services and collaboration. It realizes bi-augmentation between physical and virtual spaces using augmented reality. It also offers a context processing module to acquire, interpret and disseminate context information. In particular, the context processing module considers user's preferences and security profile for providing private and customer-oriented services. The prototype system has been implemented to support 3D animation, TTS (Text-to-Speech), augmented manual, annotation, and pre- and post-augmentation services in ubiquitous car service environments.

• ETRI Journal
• /
• v.39 no.4
• /
• pp.455-466
• /
• 2017

This study focuses on a method for sequential data augmentation in order to alleviate data sparseness problems. Specifically, we present corpus expansion techniques for enhancing the coverage of a language model. Recent recurrent neural network studies show that a seq2seq model can be applied for addressing language generation issues; it has the ability to generate new sentences from given input sentences. We present a method of corpus expansion using a sentence-chain based seq2seq model. For training the seq2seq model, sentence chains are used as triples. The first two sentences in a triple are used for the encoder of the seq2seq model, while the last sentence becomes a target sequence for the decoder. Using only internal resources, evaluation results show an improvement of approximately 7.6% relative perplexity over a baseline language model of Korean text. Additionally, from a comparison with a previous study, the sentence chain approach reduces the size of the training data by 38.4% while generating 1.4-times the number of n-grams with superior performance for English text.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.148-154
• /
• 2021

Text-Independent speaker verification needs to extract text-independent speaker embedding to improve generalization performance. However, deep neural networks that depend on training data have the potential to overfit text information instead of learning the speaker information when repeatedly learning from the identical time series. In this paper, to prevent the overfitting, we propose a segment unit shuffling layer that divides and rearranges the input layer or a hidden layer along the time axis, thus mixes the time series information. Since the segment unit shuffling layer can be applied not only to the input layer but also to the hidden layers, it can be used as generalization technique in the hidden layer, which is known to be effective compared to the generalization technique in the input layer, and can be applied simultaneously with data augmentation. In addition, the degree of distortion can be adjusted by adjusting the unit size of the segment. We observe that the performance of text-independent speaker verification is improved compared to the baseline when the proposed segment unit shuffling layer is applied.