• Korean Journal of Remote Sensing
• /
• v.27 no.4
• /
• pp.445-455
• /
• 2011

A probabilistic classification framework is presented that can combine temporal contextual information derived from an existing land-cover map in order to improve the classification accuracy of land-cover classes that can not be discriminated well when using spectral information only. The transition probability is computed by using the existing land-cover map and training data, and considered as a priori probability. By combining the a priori probability with conditional probability computed from spectral information via a Bayesian combination rule, the a posteriori probability is finally computed and then the final land-cover types are determined. The method presented in this paper can be adopted to any probabilistic classification algorithms in a simple way, compared with conventional classification methods that require heavy computational loads to incorporate the temporal contextual information. A case study for crop classification using time-series MODIS data sets is carried out to illustrate the applicability of the presented method. The classification accuracies of the land-cover classes, which showed lower classification accuracies when using only spectral information due to the low resolution MODIS data, were much improved by combining the temporal contextual information. It is expected that the presented probabilistic method would be useful both for updating the existing past land-cover maps, and for improving the classification accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2468-2487
• /
• 2015

In web searching, trustable and precise results are greatly affected by the inherent uncertainty in the input queries. Queries submitted to search engines are by nature ambiguous and constitute a significant proportion of the instances given to web search engines. Ambiguous queries pose real challenges for the web search engines due to versatility of information. Temporal based approaches whereas somehow reduce the uncertainty in queries but still lack to provide results according to users aspirations. Web search science has created an interest for the researchers to incorporate contextual information for resolving the uncertainty in search results. In this paper, we propose an Adaptive Disambiguation Approach (ADA) of hybrid nature that makes use of both the temporal and contextual information to improve user experience. The proposed hybrid approach presents the search results to the users based on their location and temporal information. A Java based prototype of the systems is developed and evaluated using standard dataset to determine its efficacy in terms of precision, accuracy, recall, and F1-measure. Supported by experimental results, ADA demonstrates better results along all the axes as compared to temporal based approaches.

• Korean Journal of Cognitive Science
• /
• v.33 no.2
• /
• pp.109-133
• /
• 2022

Episodic memory consists of a core event and the associated contexts. Although the role of the hippocampus and its neighboring regions in contextual representations during encoding has become increasingly evident, it remains unclear how these regions handle various context-specific information other than spatio-temporal contexts. Using high-resolution functional MRI, we explored the patterns of the medial temporal lobe (MTL) and cortical regions' involvement during the encoding of various types of contextual information (i.e., journalism principle 5W1H): "Who did it?," "Why did it happen?," "What happened?," "When did it happen?," "Where did it happen?," and "How did it happen?" Participants answered six different contextual questions while looking at simple experimental events consisting of two faces with one object on the screen. The MTL was divided to sub-regions by hierarchical clustering from resting-state data. General linear model analyses revealed a stronger activation of MTL sub-regions, the prefrontal lobe (PFC), and the inferior parietal lobule (IPL) during social (Who), behavioral (How), and intentional (Why) contextual processing when compared with spatio-temporal (Where/When) contextual processing. To further investigate the functional networks involved in contextual encoding dissociation, a multivariate pattern analysis was conducted with features selected as the task-based connectivity links between the hippocampal subfields and PFC/IPL. Each social, behavioral, and intentional contextual processing was individually and successfully classified from spatio-temporal contextual processing, respectively. Thus, specific contexts in episodic memory, namely social, behavior, and intention, involve distinct functional connectivity patterns that are distinct from those for spatio-temporal contextual memory.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.1052-1057
• /
• 1994

Dynamically localized self-organizing map model (DLSMM) is a new speech recognition model based on the well-known self-organizing map algorithm and dynamic programming technique. The DLSMM can efficiently normalize the temporal and spatial characteristics of speech signal at the same time. Especially, the proposed can use contextual information of speech. As experimental results on ten Korean digits recognition task, the DLSMM with contextual information has shown higher recognition rate than predictive neural network models.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.18 no.3
• /
• pp.76-88
• /
• 2015

This paper presents a classification method based on the combination of active learning with temporal contextual information extracted from past land-cover maps for the classification of crop cultivation areas. Iterative classification based on active learning is designed to extract reliable training data and cultivation rules from past land-cover maps are quantified as temporal contextual information to be used for not only assignment of training data but also relaxation of spectral ambiguity. To evaluate the applicability of the classification method proposed in this paper, a case study with MODIS time-series vegetation index data sets and past cropland data layers(CDLs) is carried out for the classification of corn and soybean in Illinois state, USA. Iterative classification based on active learning could reduce misclassification both between corn and soybean and between other crops and non crops. The combination of temporal contextual information also reduced the over-estimation results in major crops and led to the best classification accuracy. Thus, these case study results confirm that the proposed classification method can be effectively applied for crop cultivation areas where it is not easy to collect the sufficient number of reliable training data.

• Journal of Internet Computing and Services
• /
• v.16 no.6
• /
• pp.79-88
• /
• 2015

The web search engines are exposed to the issue of uncertainty because of ambiguous queries, being input for retrieving the accurate results. Ambiguous queries constitute a significant fraction of such instances and pose real challenges to web search engines. Moreover, web search has created an interest for the researchers to deal with search by considering context in terms of location perspective. Our proposed disambiguation approach is designed to improve user experience by using context in terms of location relevance with the document relevance. The aim is that providing the user a comprehensive location perspective of a topic is informative than retrieving a result that only contains temporal or context information. The capacity to use this information in a location manner can be, from a user perspective, potentially useful for several tasks, including user query understanding or clustering based on location. In order to carry out the approach, we developed a Java based prototype to derive the contextual information from the web results based on the queries from the well-known datasets. Among those results, queries are further classified in order to perform search in a broad way. After the result provision to users and the selection made by them, feedback is recorded implicitly to improve the web search based on contextual information. The experiment results demonstrate the outstanding performance of our approach in terms of precision 75%, accuracy 73%; recall 81% and f-measure 78% when compared with generic temporal evaluation approach and furthermore achieved precision 86%, accuracy 71%; recall 67% and f-measure 75% when compared with web document clustering approach.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.535-538
• /
• 1990

The main problem in speech recognition is the enormous variability in acoustic signals due to complex but predictable contextual effects. Especially in plosive consonants it is very difficult to find invariant cue due to various contextual effects, but humans use these contextual effects as helpful information in plosive consonant recognition. In this paper we experimented on three artificial neural net models for the recognition of plosive consonants. Neural Net Model I used "Multi-layer Perceptron ". Model II used a variation of the "Self-organizing Feature Map Model". And Model III used "Interactive and Competitive Model" to experiment contextual effects. The recognition experiment was performed on 9 Korean plosive consonants. We used VCV speech chains for the experiment on contextual effects. The speech chain consists of Korean plosive consonants /g, d, b, K, T, P, k, t, p/ (/ㄱ, ㄷ, ㅂ, ㄲ, ㄸ, ㅃ, ㅋ, ㅌ, ㅍ/) and eight Korean monothongs. The inputs to Neural Net Models were several temporal cues - duration of the silence, transition and vot -, and the extent of the VC formant transitions to the presence of voicing energy during closure, burst intensity, presence of asperation, amount of low frequency energy present at voicing onset, and CV formant transition extent from the acoustic signals. Model I showed about 55 - 67 %, Model II showed about 60%, and Model III showed about 67% recognition rate.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.7-21
• /
• 2024

There are various machine learning techniques such as Reinforcement Learning, Deep Learning, Neural Network Learning, and so on. In recent, Large Language Models (LLMs) are popularly used for Generative AI based on Reinforcement Learning. It makes decisions with the most optimal rewards through the fine tuning process in a particular situation. Unfortunately, LLMs can not provide any explanation for how they reach the goal because the training is based on learning of black-box AI. Reinforcement Learning as black-box AI is based on graph-evolving structure for deriving enhanced solution through adjustment by human feedback or reinforced data. In this research, for mutually exclusive decision-making, Mutually Exclusive Learning (MEL) is proposed to provide explanations of the chosen goals that are achieved by a decision on both ends with specified conditions. In MEL, decision-making process is based on the tree-based structure that can provide processes of pruning branches that are used as explanations of how to achieve the goals. The goal can be reached by trade-off among mutually exclusive alternatives according to the specific contextual conditions. Therefore, the tree-based structure is adopted to provide feasible solutions with the explanations based on the pruning branches. The sequence of pruning processes can be used to provide the explanations of the inferences and ways to reach the goals, as Explainable AI (XAI). The learning process is based on the pruning branches according to the multi-dimensional contextual conditions. To deep-dive the search, they are composed of time window to determine the temporal perspective, depth of phases for lookahead and decision criteria to prune branches. The goal depends on the policy of the pruning branches, which can be dynamically changed by configured situation with the specific multi-dimensional contextual conditions at a particular moment. The explanation is represented by the chosen episode among the decision alternatives according to configured situations. In this research, MEL adopts the tree-based learning model to provide explanation for the goal derived with specific conditions. Therefore, as an example of mutually exclusive problems, employment process is proposed to demonstrate the decision-making process of how to reach the goal and explanation by the pruning branches. Finally, further study is discussed to verify the effectiveness of MEL with experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3598-3614
• /
• 2020

With the increase of motor vehicles and tourism demand, some traffic problems gradually appear, such as traffic congestion, safety accidents and insufficient allocation of traffic resources. Facing these challenges, a model of Spatio-Temporal Dilated Convolutional Network (STDGCN) is proposed for assistance of extracting highly nonlinear and complex characteristics to accurately predict the future traffic flow. In particular, we model the traffic as undirected graphs, on which graph convolutions are built to extract spatial feature informations. Furthermore, a dilated convolution is deployed into graph convolution for capturing multi-scale contextual messages. The proposed STDGCN integrates the dilated convolution into the graph convolution, which realizes the extraction of the spatial and temporal characteristics of traffic flow data, as well as features of road occupancy. To observe the performance of the proposed model, we compare with it with four rivals. We also employ four indicators for evaluation. The experimental results show STDGCN's effectiveness. The prediction accuracy is improved by 17% in comparison with the traditional prediction methods on various real-world traffic datasets.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.3
• /
• pp.11-17
• /
• 2013

With high social demands for the diverse utilizations of forest lands, the illegal forest land use changes have increased. We studied change detection technique to detect changes in forest land use using an object-oriented segmentation of RED bands differencing in multi-temporal aerial photographs. The new object-oriented segmentation method consists of the 5 steps, "Image Composite - Segmentation - Reshaping - Noise Remover - Change Detection". The method enabled extraction of deforested objects by selecting a suitable threshold to determine whether the objects was divided or merged, based on the relations between the objects, spectral characteristics and contextual information from multi-temporal aerial photographs. The results found that the object-oriented segmentation method detected 12% of changes in forest land use, with 96% of the average detection accuracy compared by visual interpretation. Therefore this research showed that the spatial data by the object-oriented segmentation method can be complementary to the one by a visual interpretation method, and proved the possibility of automatically detecting and extracting changes in forest land use from multi-temporal aerial photographs.