• The Transactions of the Korea Information Processing Society
• /
• v.7 no.9
• /
• pp.2867-2875
• /
• 2000

In this paper, we discuss the implementation of automatic internet text categorization system. A categorization algorithm is designed and the system is implemented by back propagation learning model. Internet documents are collected according to the established categories and tested by Chi-squre ($\chi^2$) for the document leaning, and the category features are extracted. The sets of learning and separating vector are productt>d by these features. As a result of experimental evaluation, we show that this system is more improved in the performance of automatic categorization than the nearest neigbor method.

• Journal of Biosystems Engineering
• /
• v.18 no.4
• /
• pp.371-381
• /
• 1993

Threshing operation is performed by impact, compression and friction forces inside the thresher. These values should be appropriate to the crop condition to enhance the threshing and separating efficiency and to decrease the grain damage. To analyze the threshing process inside the rasp-bar type thresher, impact, friction and compression forces were measured using transducers with strain gage circuits. To measure the impact forces and friction forces between the rasp-bar and crop, full bridge strain gage circuit was built on the rasp-bar holder. To measure the compression forces and circumferential friction forces between the concave and crop, two sets of full bridge strain gage circuits were built on the T-type concave transducer. Threshing work of wheat crop with 12% of moisture content was performed at 3 levels of compression ratio and with 3 replications. Each transducer could not measure the exact forces continuously because the transducer oscillates with the forces. However they could measure maximum forces and force distribution according to the time. Average friction coefficients between crop and concave was 0.61 not showing any significant difference according to the compression ratio. Average acceleration of the crop in the cylinder appeared from $70.6m/s^2$ to $140.8m/s^2$ according to the compression ratio. The velocity of the crop at the exit of the cylinder appeared from 10.7m/s to 15.0m/s according to the compression ratio.

• Journal of Wetlands Research
• /
• v.4 no.2
• /
• pp.23-41
• /
• 2002

In this paper, changes of the coliform bacteria were investigated when aquatic plant pond was used for separating algal particles from waste stabilization ponds(WSPs). Three different types of integrated natural systems were operated. It was found that there were no significant interferences for the disinfection efficiency of two integrated systems (WSPs coupled with water hyacinth ponds) used for treating domestic sewage and upgrading the secondary effluent as well. However, when constructed wetland (CW) was combined with the shallow algal ponds and used for the secondary effluent, it seriously interfered with the disinfection efficiency due to the regrowth and/or after-growth of the coliform bacteria, which can readily metabolize the amino acids and sugars leached from plants. In order to find out the primary disinfection parameters, several sets of the batch test were run. It was found that sunlight is the most predominant factor for the coliform decay. During the night, algal toxicity partly supports the decay but during the day, deteriorates it by attenuating the sunlight. The pH in the range of 4 to 10 did not affect the decay in the dark.

• International Journal of Computer Science & Network Security
• /
• v.21 no.11
• /
• pp.294-300
• /
• 2021

The current society relies upon social media on an everyday basis, which contributes to finding which of the following supervised machine learning algorithms used in sentiment analysis have higher accuracy in detecting Malay internet slang and short forms which can be offensive to a person. This paper is to determine which of the algorithms chosen in supervised machine learning with higher accuracy in detecting internet slang and short forms. To analyze the results of the supervised machine learning classifiers, we have chosen two types of datasets, one is political topic-based, and another same set but is mixed with 50 tweets per targeted keyword. The datasets are then manually labelled positive and negative, before separating the 275 tweets into training and testing sets. Naïve Bayes and Random Forest classifiers are then analyzed and evaluated from their performances. Our experiment results show that Random Forest is a better classifier compared to Naïve Bayes.

• Journal of Broadcast Engineering
• /
• v.22 no.5
• /
• pp.632-642
• /
• 2017

In this paper, we propose a DNN based speech detection system using acoustic characteristics and context information of media audio. The speech detection for discriminating between speech and non-speech included in the media audio is a necessary preprocessing technique for effective speech processing. However, since the media audio signal includes various types of sound sources, it has been difficult to achieve high performance with the conventional signal processing techniques. The proposed method improves the speech detection performance by separating the harmonic and percussive components of the media audio and constructing the DNN input vector reflecting the acoustic characteristics and context information of the media audio. In order to verify the performance of the proposed system, a data set for speech detection was made using more than 20 hours of drama, and an 8-hour Hollywood movie data set, which was publicly available, was further acquired and used for experiments. In the experiment, it is shown that the proposed system provides better performance than the conventional method through the cross validation for two data sets.

• The Korean Journal of Applied Statistics
• /
• v.18 no.2
• /
• pp.421-434
• /
• 2005

Machine learning has been extensively studied in recent years as effective tools in pattern classification problem. Although there have been several approaches to machine learning, we focus on the mathematical programming (in particular, multi-objective and goal programming; MOP/GP) approaches in this paper. Among them, Support Vector Machine (SVM) is gaining much popularity recently. In pattern classification problem with two class sets, the idea is to find a maximal margin separating hyperplane which gives the greatest separation between the classes in a high dimensional feature space. However, the idea of maximal margin separation is not quite new: in 1960's the multi-surface method (MSM) was suggested by Mangasarian. In 1980's, linear classifiers using goal programming were developed extensively. This paper proposes a new family of SVM using MOP/GP techniques, and discusses its effectiveness throughout several numerical experiments.

• Parasites, Hosts and Diseases
• /
• v.47 no.4
• /
• pp.353-357
• /
• 2009

Cryptosporidium can cause gastrointestinal diseases worldwide, consequently posing public health problems and economic burden. Effective techniques for detecting contaminated oocysts in water are important to prevent and control the contamination. Immunomagnetic separation (IMS) method has been widely employed recently due to its efficiency, but, it is costly. Sucrose floatation technique is generally used for separating organisms by using their different specific gravity. It is effective and cheap but time consuming as well as requiring highly skilled personnel. Water turbidity and parasite load in water sample are additional factors affecting to the recovery rate of those 2 methods. We compared the efficiency of IMS and sucrose floatation methods to recover the spiked Cryptosporidium oocysts in various turbidity water samples. Cryptosporidium oocysts concentration at 1, $10^1$, $10^2$, and $10^3$ per $10{\mu}l$ were spiked into 3 sets of 10 ml-water turbidity (5, 50, and 500 NTU). The recovery rate of the 2 methods was not different. Oocyst load at the concentration < $10^2$ per 10 ml yielded unreliable results. Water turbidity at 500 NTU decreased the recovery rate of both techniques. The combination of sucrose floatation and immunofluorescense assay techniques (SF-FA) showed higher recovery rate than IMS and immunofluorescense assay (IMS-FA). We used this SF-FA to detect Cryptosporidium and Giardia from the river water samples and found 9 and 19 out of 30 (30% and 63.3%) positive, respectively. Our results favored sucrose floatation technique enhanced with immunofluorescense assay for detecting contaminated protozoa in water samples in general laboratories and in the real practical setting.

• Economic and Environmental Geology
• /
• v.26 no.2
• /
• pp.227-237
• /
• 1993

Lineaments in the Kyungsang basin most intensely develop in the East coast domain including the Yangsan fault, which dominantly run in NNE direction. The geometry of small fault population near or along the Yangsan fault represents the dominant strikes of N35E, high angle dips and shallowly plunging rakes with dextral movement sense. Stereographic solution on the Yangsan fault geometry gives the dip of 88SE, the slip direction of 17,024 and the slip rake of 18, which were determined from the strike (N23E) of the fault measured on map, and the average attitude (N35E, 84SE) and fault striation (16, 037) of small fault population considered as Riedel shears. It is judged from the geometry of small fault population to the main Yangsan fault and dragging features of bedding attitude near the fault that the Yangsan fault was produced from dextrally strike-slip movement. The movement of the Yangsan and the adjacent parallel faults is thought to be taken place much later than the other fault sets in the Kyungsang basin. It might occur during the geologic age from Eocence to early Miocene according to the consideration of K-Ar ages of the igneous rocks near the fault. The estimated paleostress state indicates ENE shortening and NNW extension. The displacement of the Yangsan fault in the study area is not constant along the fault but decreases from the south to the north. Taking the northern end of the study area as a separating point the whole extension of the Yangsan fault may be divided into southern and northern segments.

• Journal of Animal Science and Technology
• /
• v.63 no.4
• /
• pp.740-750
• /
• 2021

This study was performed to investigate genetic relationship of age at first calving (AFC) with body development and reproduction capacity in female Hanwoo. The data sets of 52,299 reproduction records of 19,566 heads calved from 2011 to 2019 and conformation records of 19565 heads born from 1997 to 2017 were analyzed by separating them based on age at first calving. In the analysis, conformation traits included the body condition score (BCS), height (HT), body width (BW), hip bone length (HL), hip bone width (HW), and final score (FS). The heritability and genetic correlation of AFC with conformation traits and calving interval (CI) were analyzed. The heritability for AFC was 0.716, 0.087, and 0.097 for the before and after 24 months and combined data, respectively. The genetic correlation of AFC before 24 months with BCS was -0.778, -0.600 with BW, and 0.442 with HT. The genetic correlation of AFC after 24 months with HT was -0.826, -0.706 with BW, -0.623 with HL, -0.456 with HW, and -0.675 with FS. When the first calving age of young heifers approached 24 months, BCS and BW decreased, and HT increased. When first calving is delayed to after 24 months, the conformation traits become smaller, which indicates that conformation to some extent affects the delay in AFC. The genetic correlation between CI and AFC was -0.116, 0.307, and 0.250 for the before and after 24 month of AFC and combined data, respectively. When first calving date approached 24 months, CI was reduced. The obtained results suggest that it is important that first calving occurs at an appropriate age. Additional research is needed to perform proper genetic evaluation of first calving age in Hanwoo cattle in Korea.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.1957-1980
• /
• 2021

The significant aim of this research has always been to evaluate the mechanism for efficient and inherently aware usage of vitality in-home devices, thus improving the information of smart metering systems with regard to the usage of selected homes and the time of use. Advances in information processing are commonly used to quantify gigantic building activity data steps to boost the activity efficiency of the building energy systems. Here, some smart data mining models are offered to measure, and predict the time series for energy in order to expose different ephemeral principles for using energy. Such considerations illustrate the use of machines in relation to time, such as day hour, time of day, week, month and year relationships within a family unit, which are key components in gathering and separating the effect of consumers behaviors in the use of energy and their pattern of energy prediction. It is necessary to determine the multiple relations through the usage of different appliances from simultaneous information flows. In comparison, specific relations among interval-based instances where multiple appliances use continue for certain duration are difficult to determine. In order to resolve these difficulties, an unsupervised energy time-series data clustering and a frequent pattern mining study as well as a deep learning technique for estimating energy use were presented. A broad test using true data sets that are rich in smart meter data were conducted. The exact results of the appliance designs that were recognized by the proposed model were filled out by Deep Convolutional Neural Networks (CNN) and Recurrent Neural Networks (LSTM and GRU) at each stage, with consolidated accuracy of 94.79%, 97.99%, 99.61%, for 25%, 50%, and 75%, respectively.