Browsing by Author "Pao, Hsing-Kuo"

Now showing 1 - 4 of 4

A Unified Approach on Active Learning Dual Supervision
(2019-01-01)
Chriswanto, Adrian
;
Pao, Hsing-Kuo
;
Leet, Yuh-Jye
;
應用數學系
;
Department of Applied Mathematics
Active Learning (AL) is a machine learning framework that aims to efficiently select a limited labeled data to construct an effective model given huge amount of unlabeled data on the side. Most studies in AL focus on how to select the unlabeled data to be labeled by a human oracle in order to maximize the performance gain of the model with as little labeling effort as possible. In this work, however, we focus not only on how to select appropriate data instances but also how to select informative features, more specifically, categorical features to be labeled by the oracle in a unified manner. The unification means that we select the best choice of item to label where the item can be either a feature or an instance on each iteration given a unified scoring function to make the decision. The method that we propose is by synthesizing new instances that represent a set of features. By utilizing synthesized instances, we can treat this set of features as if they are regular instances. Therefore they could be compared on an equal ground when the model tries to select which instances to be labeled by the oracle. The features used to build the synthesized instances need to be carefully selected so the resulting synthesized instances could improve the model and not introducing any contradicting information. We utilize hierarchical clustering in order to group features that own similar content. This is done first by picking clusters whose label purity are estimated to be high. Then we score a feature based on how common the feature is in the cluster and how related the feature is to the estimated majority label. The top scoring features will then be used to synthesize instances. We demonstrate the effectiveness of the proposed method through a few data sets that consist of only categorical features where the feature labeling makes more sense to labeling oracles. The experiment results show that adopting the unified approach creates clear benefit to model construction, especially in the early stage where we can efficiently obtain an effective model through only a few iterations, compared to the one using only instance labeling for model construction.
Big Active Learning
(2017-01-01)
Huang, Er-Chen
;
Pao, Hsing-Kuo
;
Lee, Yuh-Jye
;
應用數學系
;
Department of Applied Mathematics
Active learning is a common strategy to deal with large-scale data with limited labeling effort. In each iteration of active learning, a query is ready for oracle to answer such as what the label is for a given unlabeled data. Given the method, we can request the labels only for those data that are essential and save the labeling effort from oracle. We focus on pool-based active learning where a set of unlabeled data is selected for querying in each run of active learning. To apply pool-based active learning to massive high-dimensional data, especially when the unlabeled data set is much larger than the labeled set, we propose the APRAL and MLP strategies so that the computation for active learning can be dramatically reduced while keeping the model power more or less the same. In APRAL, we avoid unnecessary data re-ranking given an unlabeled data selection criteria. To further improve the efficiency, with MLP, we organize the unlabeled data in a multi-layer pool based on a dimensionality reduction technique and the most valuable data to know their label information are more likely to store in the top layers. Given the APRAL and MLP strategies, the active learning computation time is reduced by about 83% if compared to the traditional active learning ones; at the same time, the model effectiveness remains.
Compressed Learning for Time Series Classification
(2016-01-01)
Lee, Yuh-Jye
;
Pao, Hsing-Kuo
;
Shih, Shueh-Han
;
Lin, Jing-Yao
;
Chen, Xin-Rong
The time series classification has been studied for various applications in the last decades. In the time series classification problem, we decide the class information based on a small piece of the time series inputs. In general, the approaches to time series classification can be categorized into three types, distance-based, model-based, and feature-based approaches. In this research, we focus on the feature-based methods, which represent time series as a set of characterized values. It is quite often the case that features generated by existing representation techniques are not transparent to domain experts and the feature that are selected for classification are not completely interpretable. We aim to propose a novel time series representation, called Envelope to solve the problem. The proposed supervised feature extraction method transforms time series into simple 1/0/-1 values. A heuristic is introduced to determine the most appropriate representation which includes the features that are the best to discriminate data of different labels. Moreover, this new representation enjoys the characteristic of sparsity which is an essential property when we need to apply compressed sensing techniques. With this advantage, we can benefit from high transmission efficiency, the reduction of required storage and model complexity. We conduct a series of tests on various benchmark time series data to show the effectiveness of the proposed method. Other than the classification effectiveness, we demonstrate how to visualize the similarity between time series of the same and different kinds from the proposed Envelope method.
Dealing with Interleaved Event Inputs for Intrusion Detection
(2019-01-01)
Pao, Hsing-Kuo
;
Lee, Fong-Ruei
;
Lee, Yuh-Jye
;
應用數學系
;
Department of Applied Mathematics
We propose an intrusion detection method that can deal with interleaved event inputs. The event sequences may be alert sequences in a network or running processes on a host which are both considered to contain mixed behaviors with unpredictable orders in the temporal domain. To detect intrusions with interleaved event sequences, one of the major difficulties is to separate the interleaved events that are produced by different users or for different intentions. We propose a novel ATM algorithm to extract subsequences that characterize different behaviors; afterwards, a method that is based on graph representation is used to detect intrusions. In a network, there could be intruders who plan a DDoS attack on an environment that has mostly benign users. The proposed method can distinguish between different pieces of network data that represent different behaviors and locate where the intrusion is. On a host, users without enough privilege may inappropriately gain access to data that they are not supposed to see. The proposed method can detect the event subsequence that is associated with the unauthorized activity given a usage sequence from users such as process, command or log sequences. Given the network or host-based data, the experiment results show that the proposed method can reach high precision and recall rates at the same time in the intrusion detection task. Moreover, the graphs produced by the proposed ATM method are also compared to the graphs generated from other methods to confirm that the ATM-based graph representation indeed describes meaningful transitions between events.