• Journal of the Korea Society of Computer and Information
• /
• v.25 no.10
• /
• pp.87-94
• /
• 2020

Social reviews such as SNS feeds and blog articles have been widely used to extract keywords reflecting opinions and complaints from users' perspective, and often include proper nouns or new words reflecting recent trends. In general, these words are not included in a dictionary, so conventional morphological analyzers may not detect and extract those words from the reviews properly. In addition, due to their high processing time, it is inadequate to provide analysis results in a timely manner. This paper presents a method for efficient keyword extraction from social reviews based on the notion of cohesion scoring. Cohesion scores can be calculated based on word frequencies, so keyword extraction can be performed without a dictionary when using it. On the other hand, their accuracy can be degraded when input data with poor spacing is given. Regarding this, an algorithm is presented which improves the existing cohesion scoring mechanism using the structure of a word tree. Our experiment results show that it took only 0.008 seconds to extract keywords from 1,000 reviews in the proposed method while resulting in 15.5% error ratio which is better than the existing morphological analyzers.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.5
• /
• pp.567-575
• /
• 2001

This paper suggests new method to get fixed size parameter from different length of voice signals. The efficiency of speech recognizer is determined by how to compare the similarity(distance of each pattern) of the parameter from voice signal. But the variation of voice signal and the difference of speech speed make it difficult to extract the fixed size parameter from the voice signal. The method suggested in this paper is to normalize the parameter at fixed size by using the 2 dimension DCT(Discrete Cosine Transform) after representing the parameter by spectrogram. To prove validity of the suggested method, parameter extracted from 32 auditory filter-bank(it estimates auditory nerve firing probabilities) is used for the input of neural network after being processed by 2 dimension DCT. And to compare with conventional methods, we used one of conventional methods which solve time alignment problem. The result shows more efficient performance and faster recognition speed in the speaker dependent and independent isolated word recognition than conventional method.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.2 s.344
• /
• pp.51-59
• /
• 2006

Von Neumann architecture suffers from the interface between the central processing unit and the memory, which is called 'Von Neumann bottleneck' In this paper, we propose a scalable general-purpose associative processor (AP) based on content-addressable memory (CAM) which solves this problem and is suitable for the search-oriented applications. We propose an efficient instruction set and a structural scalability to extend for larger applications. We define twelve instructions and provide some reduced instructions to speed up which execute two instructions in a single instruction cycle. The proposed AP performs in a bit-serial, word-parallel fashion and can be considered as a 32-bit general-purpose parallel processor with a massively parallel SIMD structure. We design and simulate a maximum/minumum search greater-than/less-than search, and parallel addition to verify the proposed architecture. The algorithms are executed in a constant time O(k) regardless of the number of input data.

• Journal of KIISE
• /
• v.44 no.1
• /
• pp.57-62
• /
• 2017

Traditional Korean morphological analysis and POS tagging methods usually consist of two steps: 1 Generat hypotheses of all possible combinations of morphemes for given input, 2 Perform POS tagging search optimal result. require additional resource dictionaries and step could error to the step. In this paper, we tried to solve this problem end-to-end fashion using sequence-to-sequence model convolutional features. Experiment results Sejong corpus sour approach achieved 97.15% F1-score on morpheme level, 95.33% and 60.62% precision on word and sentence level, respectively; s96.91% F1-score on morpheme level, 95.40% and 60.62% precision on word and sentence level, respectively.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.15 no.4
• /
• pp.26-32
• /
• 2016

An autonomous car(driverless car, self-driving car, robotic car) is a vehicle that is capable of sensing its environment and navigating without human input. Autonomous cars can detect surroundings using a variety of techniques such as radar, lidar, GPS, odometry, and computer vision. Advanced control systems interpret sensory information to identify appropriate navigation paths, as well as obstacles and relevant signage. Autonomous cars have control systems that are capable of analyzing sensory data to distinguish between different cars on the road, which is very useful in planning a path to the desired destination. An issue tree, also called a logic tree, is a graphical breakdown of a question that dissects it into its different components vertically and that progresses into details as it reads to the right.Issue trees are useful in problem solving to identify the root causes of a problem as well as to identify its potential solutions. They also provide a reference point to see how each piece fits into the whole picture of a problem. Using Issue-Tree menthods, transportation safety policies were developed with autonompus vehicle in mind.

• Journal of Digital Contents Society
• /
• v.19 no.2
• /
• pp.351-362
• /
• 2018

This study investigated how Schwartz' value system(1992) would interfere with SNS users' motivations and behavioral responses. The study result shows that values are characterized in terms of openness, mutual reciprocity, self enhancement, normative compliance, and security. Each of them exerted differential impact on SNS usage motives, social capital, ad response, and word-of-mouth, among others. The five values were used as an input for segmenting SNS users and clustering method produced four value segments; experience seeker, interdependent sympathizer, self enhancer, and norm-bound. Each value group not only influenced SNS perceptions and behavioral responses differently, but also showed a systematic relationship with SNS service types. The study findings demonstrate that Schwartz's value system provides a very useful theoretical basis for understanding the psychological mechanism underlying SNS usage.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.11
• /
• pp.33-40
• /
• 2021

Due to advanced complex strategies, the complexity of information that a commander must analyze is increasing. An intelligent service that can analyze battlefield is needed for the commander's timely judgment. This service consists of extracting knowledge from battlefield information, building a knowledge base, and analyzing the battlefield information from the knowledge base. This paper extract information similar to an input query by embedding the knowledge base built in the 2^nd step. The transformation model is needed to generate the embedded knowledge base and uses the random-walk algorithm. The transformed information is embedding using Word2Vec, and Similar information is extracted through cosine similarity. In this paper, 980 sentences are generated from the open knowledge base and embedded as a 100-dimensional vector and it was confirmed that similar entities were extracted through cosine similarity.

• Smart Media Journal
• /
• v.7 no.4
• /
• pp.17-23
• /
• 2018

We proposed a LSTM-based RNN model that can effectively perform the automatic spacing characteristics. For those long or noisy sentences which are known to be difficult to handle within Neural Network Learning, we defined a proper input data format and decoding data format, and added dropout, bidirectional multi-layer LSTM, layer normalization, and attention mechanism to improve the performance. Despite of the fact that Sejong corpus contains some spacing errors, a noise-robust learning model developed in this study with no overfitting through a dropout method helped training and returned meaningful results of Korean word spacing and its patterns. The experimental results showed that the performance of LSTM sequence-to-sequence model is 0.94 in F1-measure, which is better than the rule-based deep-learning method of GRU-CRF.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.837-852
• /
• 2021

Recurrent neural network (RNN) architectures have been used for language modeling (LM) tasks that require learning long-range word or character sequences. However, the RNN architecture is still suffered from unstable gradients on long-range sequences. To address the issue of long-range sequences, an attention mechanism has been used, showing state-of-the-art (SOTA) performance in all LM tasks. A differentiable neural computer (DNC) is a deep learning architecture using an attention mechanism. The DNC architecture is a neural network augmented with a content-addressable external memory. However, in the write operation, some information unrelated to the input word remains in memory. Moreover, DNCs have been found to perform poorly with low numbers of weight parameters. Therefore, we propose a robust memory deallocation method using a limited retention vector. The limited retention vector determines whether the network increases or decreases its usage of information in external memory according to a threshold. We experimentally evaluate the robustness of a DNC implementing the proposed approach according to the size of the controller and external memory on the enwik8 LM task. When we decreased the number of weight parameters by 32.47%, the proposed DNC showed a low bits-per-character (BPC) degradation of 4.30%, demonstrating the effectiveness of our approach in language modeling tasks.

• Journal of Convergence for Information Technology
• /
• v.12 no.2
• /
• pp.47-53
• /
• 2022

Recently, with the change of the intelligent security paradigm, study to apply various information generated from various information security systems to AI-based anomaly detection is increasing. Therefore, in this study, in order to convert log-like time series data into a vector, which is a numerical feature, the CBOW and Skip-gram inference methods of deep learning-based Word2Vec model and statistical method based on the coincidence frequency were used to transform the published ADFA system call data. In relation to this, an experiment was carried out through conversion into various embedding vectors considering the dimension of vector, the length of sequence, and the window size. In addition, the performance of the embedding methods used as well as the detection performance were compared and evaluated through GRU-based anomaly detection model using vectors generated by the embedding model as an input. Compared to the statistical model, it was confirmed that the Skip-gram maintains more stable performance without biasing a specific window size or sequence length, and is more effective in making each event of sequence data into an embedding vector.