Search

This study explored the usability of Near Field Communication (NFC) as a contactless data transmission in an airport to help passenger and baggage management run efficiently, passenger use their smartphone to do frequent authorization in airport and use an NFC tag to store their baggage status and location during the process in airport. The prototype system was developed using several NFC readers and applications developed on Android-based phones. The test results show that the prototype can work according to the design.

Voting is an activity organized within a country on which the population is obliged to exercise the right to vote and to be elected. Voting is generally done in the traditional way, how the inhabitants go directly to the voting booths, but there are also some countries and local governments that have polled electronically using information technology or called e-voting. The voting has become an important part of the democratic system. Poll result data can show effectively in predicting election results. AHP is one of the techniques used to support decision making. In this paper we used AHP algorithm modification with Double-track - Most Significant Operation First (DT-MSOF) concept to help determine voter candidate recommendation during polling. AHP algorithm modification is performed in order to reduce the number of operations that need to be done, where based on the results of experiments the use of the concept of DT-MSOF on AHP can reduce the number of operations that need to be done to get the results (decisions) similar to that obtained by conventional AHP. In this paper we have applied AHP algorithm modification on candidate selection polls compared with AHP algorithm, where the result of poll recommendation for candidate selection shows the same result and has succeeded to reduce the number of operation.

Verifiability is one of the parameters in e-voting that can increase confidence in voting technology with several parties ensuring that voters do not change their votes. Voting has become an important part of the democratization system, both to make choices regarding policies, to elect representatives to sit in the representative assembly, and to elect leaders. the more voters and the wider the distribution, the more complex the social life, and the need to manage the voting process efficiently and determine the results more quickly, electronic-based voting (e-Voting) is becoming a more promising option. The level of confidence in voting depends on the capabilities of the system. E-voting must have parameters that can be used as guidelines, which include the following: Accuracy, Invulnerability, Privacy and Verifiability. The implementation of the simple verifiability metric to measure the degree of verifiability in the e-voting protocol, the researchers can calculate the degree of verifiability in the e-voting protocol and the researchers have been able to assess the proposed e-voting protocol with the standard of the best degree of verifiability is 1, where the value of 1 is is absolutely verified protocol.

The growth of technology aims to help human’s activity. One of human’s activity which could use technology is in the transportation area by implementing machine learning. This paper discusses the semi-ensemble method for classifying traffic condition, which could be used to classify the traffic condition for shorten travel time in the road. Semi-ensemble that applied is using voting system which consists of several neural networks. The proposed method in this paper gives better performance result than single neural network Even though the performance result is not increased significantly, enhancement in semi-ensemble with voting system which comes from best-5 performance neural networks also give better result than voting system using 10 neural networks. The performance increased from 82.58% to 82.81% for its accuracy and the rests of performance value increased from 65.09% to 65.62%.

When managing Real-Time System execution, dynamic deadlines are problems that are not easy to handle. If a real-time computation is built using the traditional approach (Run-to-Completion), then there is no tolerance for the time of completion of the execution. Using the MSOF-DT computation strategy, the RT task can be more tolerant of the reduced processing time (deadline changes). Besides that, in this paper, we also propose a real-time process scheduling algorithm based on the MSOF-DT strategy, called EDSEF/MSOF-DT. The schedulability of the schedule produced by EDSEF/MSOF-DT can be better than the schedule generated by the EDF algorithm and also higher than using (Min_D+ Min_C) policy alone without MSOF-DT, both in cases without interruptions or cases that contain interruptions.

Alignment recovery is the process of recovering optimal alignment results based on weighted computational scores. Alignment recovery in classical dynamic programming (DP) is obtained by backtracking a trace created in the computational scoring phase. Bit parallelism is a parallel computing technique done by increasing the size of the word processing unit. The actual value of the score in the DP matrix is replaced by the difference of the adjacent matrix cell values. This requires adjustment of the scoring computation and its alignment recovery mechanism, especially for integer weights. General integer scoring involves a set of integers as its weights in bit-parallelism score computation that are not yet accompanied by alignment recovery. This paper presents an algorithm to recover the final alignment results from interpreting the horizontal and vertical adjacent-cell bit parallelism score values. The results show that the algorithm produces the same match pair result as the classical backtracking DP algorithm with less gap insertion. The algorithm requires O(m) space and time.

Dynamic Time Warping is one algorithm that is often used in analyzing the similarity of time series data. One disadvantage of using DTW is its large time complexity. In this paper, the MSOF-DT computing strategy is used to modify an algorithm to check the similarity of timeseries previously based on DTW to be based on MSOF-DTW. Theoretical analysis shows that MSOF-DTW has a lower limit of time complexity that is better than DTW. Simulation results prove that MSOF-DTW is better than DTW in terms of the number of operations, execution time, productivity, and accomplished tasks.

Metabolic pathways are important biochemical processes in living organisms. The metabolism is a series of enzyme-catalyzed biochemical processes. The presence of enzymes can be used as a marker for the presence of a metabolic process. The functions of metabolism can be derived from the similarity of the catalystic protein sequences. HMM gives reasonably good prediction in catalyst prediction. The paper presents a brief description how a prototype for metabolic pathway prediction can be built using HMM.

Heterogeneous computing has been one of the mainstreams in the area of high-performance computing. One type of heterogeneous computer is a computer with CPUs, GPUs, and is programmed using OpenCL software framework. To achieve its best performance, an OpenCL kernel should be executed on the right processor. However, the OpenCL framework requires programmers to select processors for their workload. Thus, programmers need to understand which processor is best for their OpenCL kernels. If programmers did not aware about their kernels and the capabilities available processors, then the kernels are potentially not executed on the best processor. To overcome the problem, this research proposes an approach that selects the best processor to execute an OpenCL kernel. The processor selection is accomplished by k-nearest neighbor (KNN) algorithm. In order to achieve acceptable accuracy, the features to be used with KNN are selected beforehand. The selection of features involves two models, namely filter model, and wrapper model. This research uses n-fold cross-validation to evaluate the performance of the proposed method. According to the experiments, on selecting best processors for workloads on four processors, the approach achieved the accuracy of 91.5% to 100%. The result of the experiments also concludes that the features with the most significant contribution are vector integer operations and vector global memory access, or vector integer operations and vector global memory access.

Consultation-based Collaborative Intrusion Detection Network (CIDN) aims to improve the detection accuracy of an Intrusion Detection System (IDS). Acquaintance management has become a vital process which responsible for the acquaintance selection in CIDN. In an acquaintance management algorithm, the trustworthiness of the IDS is estimated by the used of risk-cost analysis. However, the inaccurate estimation of the risk cost value has lead to a poor acquaintance list quality. As there is more than one risk-cost analysis, it is beneficial to analyze the used of different risk-cost analysis in an acquaintance management algorithm. We implement three different risk-cost analysis in four different environments of CIDN. The result shows that in the mix component of the DDoS attack, our proposed risk-cost analysis provide higher acquaintance detection accuracy.

The need to manage the voting process more efficiently, more flexible access, and faster results setting make e-voting systems a promising alternative. ThreeBallot was proposed to overcome the contradiction among some e-voting requirements, i.e. anonymity, privacy, and verifiability. Verifiability is an electronic voting system should provide a method to verify that the system works as expected. The verification of the original ThreeBallot method we call partial verification, because the voter only has a 1/3 belief that his voice exists and does not change. By making some modifications and additions to the ThreeBallot system, a system which allows for full verification is made (it is assured that three parts of the ballot do exist and remain unchanged). Unfortunately, by using original ThreeBallot, voter has only partial confidence in verifying his/her vote. We propose a modified ThreeBallot-based e-voting system to provide full verification without sacrificing either anonymity or privacy for verifiability notion.

Distributed Denial of Service (DDoS) attack growing larger as growing advance of internet technology and scale. Traffic Anomaly-based Intrusion Detection/Prevention System (IDS/IPS) has proposed as its capability to detect DDoS attack. Unfortunately, the anomaly detection algorithm was limited to detect the existence of an attack. To determine the type of attack, the system must occupy multiclass classification algorithm. However, the complexity of distance-based classification algorithm is always higher than detection algorithm. This research view IPS as one integrated sequence of detection, classification, and prevention process. We propose a multi-stage IPS with the used of distance-based anomaly detection analysis in the first stage, and classification analysis in later stages. The simulation result shows lower processing time of the IPS, with a capability of choosing a certain action for certain type of DDoS attack.

In Real-Time Systems, both logical requirements and temporal constraints must be satisfied. With the assumption that logical correctness of the computation is already not a big issue anymore, this paper surveys existing approaches to reduce computation time. In general, the computation results are available only after all computation elements (or operations) have been executed. Furthermore, existing approaches are not paying much attention to the result's accuracy incremental for each execution phase. We propose a new approach which consists of the concept of intermediate results, sorting the computation elements by its significance, and the concept of interval arithmetic to decide whether the next computation element is necessary to be executed or not. Using the sum of product function model as the case study, our simulation shows that this new approach gives a better accuracy rate improvement (for each execution phase) and an ability to lessen the number of executed computation needed to get the correct final result (which implies to the reduction in computation time).

Grid scheduling has been intensively studied, tools that enable the simulation of grids have been developed. These simulation tools have the common objective of developing and evaluating scheduling algorithms for grids. By using the simulation tools, knowledge about the behaviour of scheduling system will be more easily understood. However, knowledge about the method and how to use the tools is required. This paper will describe the use of SimDAG to simulate MinMin, MaxMin and HEFT scheduling algorithms in HPC.

Anomaly traffic detection is one method to detect attack in internet, especially Distributed Denial of Service (DDoS). Here, traffic analysis which reveal the observed traffic pattern is key important process as the anomaly decision was taken based on traffic analysis. This research analyzed several statistical analysis of traffic datasets categorized as normal, DDoS attack and flashcrowd. Analysis done in aggregate and per-flow traffic level which showed similarity and difference in each category. Windowing technic used to measure instantaneous statistical value. The result showed several statistical difference which could be used to categorized types of anomaly, especially to identify normal threshold. Flow to normal average flow distance distribution perform not followed Gaussian distribution.

Flashcrowd and Distributed Denial of Service (DDoS) almost has similar symptom across network and server. But security element such Intrusion Detection System (IDS) must handle both differently. If IDS cannot differentiate flashcrowd and DDoS attack, Quality of Service of legal user traffic in flashcrowd will degraded. So it is important for IDS to differentiate between flashcrowd and DDoS. Many earlier comparison method could sense the anomalous event, but not pay much attention to identify which flow was the anomaly. We presented residual calculation between windowed aggregate traffic statistical value combination. With residual calculation among statistical percentile 10th and mean, a high accuracy of flashcrowd and DDoS differentiation of synthetic anomalous event gained. This method could directly identify the anomalous flow and perform visual analysis to detect the start to end of both event.

This study is concerned with personnel availability in meeting as a factor that can influence a meeting's value. Since a meeting schedule is not guaranteed to be fixed until the due date, this study proposes an approach to handling meeting scheduling conflict automatically ahead of the meeting. It specifically aims to maintain a meeting schedule through meeting negotiation strategies in such a way that the failure to fulfill the meeting quorum can be decreased; in other words, rescheduling phase can be avoided. A meeting schedule in this study is a timeslot selected by calculating the social welfare of personnel availability using the Clarke Tax Mechanism. There are two cases which are presented here as the basis for the negotiation process: the negotiation strategy for personnel with zero tax and the negotiation strategy for personnel who are taxed by some costs. The strategies which include personnel replacement and quorum threshold relaxation are embedded into personnel tax criteria. By using these meeting scheduling negotiation strategies, personnel availability can be maintained in order to satisfy the meeting constraint of personnel minimum attendance.

A device that can help the driver so that the vehicle does not exit lanes known as Lane Keeping Assist System (LKAS). LKAS useful to help the driver, especially on roads with monotonous situations such as toll roads. This paper present the development of fuzzy logic-based algorithm for LKAS. Remote control (RC) car is used as a tool to test the algorithm in real time situation. The test results at three different speeds are slow, medium, and high indicate that the algorithm can work according to design.

Time-to-market pressure and productivity gap are two factors that encourage the Electronic Design Automation (EDA) industry and researcher of embedded system to enhanche embedded system design method. Current embedded system design approach, Register Transfer Level, is not sufficient to fulfill the embedded system design necessity. It needs a new design method above RTL, higher abstraction layer. Electronic System Level is general term that used to name system above RTL. ESL definition so far is a system above RTL including hardware and software. In this paper we designed a Transaction Level Modeling for early verification on embedded system design. This modeling is used to know functionality fulfillment at early stage. We specify four kind model for early verification purpose.

Along with the learning convergence and nonstationary equilibria, important thing to be solved in Reinforcement Learning is the slow-learning problem. In highly dynamic and stochastic systems, Markov Decision Processes is often used to model the situation, and RL is used to produce optimum control values which are expressed by optimum policy. However, approaches to solving MDP’s using RL depend on storing the optimal value function or Q-value function and action models as tables do not scale to large state-spaces. The number of states grow exponentially along with the number of agents, the state and action space, which cause the learning process become very slow since it needs a very large amount of computer memory, and yet it needs more computation than the most computer performance nowadays have offered. This paper addressed curse of dimensionality problem by reducing the states in the MDP iteratively using a novel algorithm called State Thresholding in Reinforcement Learning (STRL). STRL accelerate the learning process and empirically proven to outperformed Q learning algorithm.

Researches have been conducted to overcome Distributed Denial of Service (DDoS) flooding attack. Beside the use of signature based detection, anomaly based detection is also used to detect the attack. Several methods such as statistic, information theory, data mining and forecasting have been proposed. In several researches, they just focused to detect the traffic anomaly, but not to recognize the types of anomaly that were detected such as flashcrowd, types of botnet, types of DDoS, and prevention action. In this paper we categorize anomaly traffic detection system based on process and capability focus. Anomaly detection system process including traffic features, preprocessing, and detection process. Capability focus based on each main research problem to be solved, there are detectingonly anomaly, types of anomaly, and prevention system that include process to overcome the attack. At the end of paper, we provide overview of research direction and opportunities that may be done in future research.

The process of multiplications in Finite Fields required huge resources. If those implemented in the Elliptic Curve Cryptography (ECC), the need of resources would be inflated because those processes were enough to dominated in every ECC level. There were many researches to found methods that could reduce the number of multiplications. One method that was well-known and developed was Karatsuba ofman algorithm, where the development of this research were General Karatsuba Multiplier, Efficient Multiplier in GF((2n)4) Generalizations of Karatsuba Algorithm, 5-6-7 Term Karatsuba-Like Formulae, and improved modulo functions. From those researches is made a better algorithm than Generalizations of Karatsuba Algorithm and it is expected to leads to generalize the form n-Term Karatsuba-Like Formulae.

Big data phenomenon arises from the increasing number of data collected from various sources, including the internet. Big data is not only about the size or volume. Big data posses specific characteristics (volume, variety, velocity, and value - 4V) that make it difficult to manage from security point of view. The evolution of data to become big data rises another important issues about data security and its management. NIST defines guide for conducting risk assessments on data, including risk management process and risk assessment. This paper looks at NIST risk management guidance and determines whether the approach of this standard is applicable to big data by generally define the threat source, threat events, vulnerabilities, likelihood of occurence and impact. The result of this study will be a general framework defining security management on Big data.

Load balancing is a major problem in grid computing systems to balance the computing resources are used to achieve optimal load processing time . The method used in load balancing by adjusting burden on computing resources that will be provided . Coordination of each computing resource is done by looking at the ability of each computing resource . Load balancing algorithm for heterogeneous computing resources in a grid computing environment using the integration of the two heuristic optimization algorithm Genetic Algorithm (GA) and Tabu Search (TS) as the Integration of Genetic and Tabu Search (IGTS) . IGTS algorithm utilizes the strengths of each algorithm used , the first force in the GA global search and local search in both the strength of the TS method . IGTS algorithm will map the jobs to be processed on the computational resources that are used to get a faster time makespan and the minimum value of balance.

In this paper, we present the unsupervised Part-of-Speech (PoS) induction algorithm to improve translations quality on statistical machine translation. The proposed algorithm is an extension of the algorithm Word-Similarity-Based (WSB) clustering. In the clustering, the similarity between words is measured by its grammatical relation with other words. The grammatical relation is represented as the n-gram relation. We extend the WSB clustering by take into account for the previous words in measuring the grammatical relation. The clustering results are then used in the English-Indonesia statistical machine translation. The experiments were conducted using MOSES as the machine translation decoder, and were evaluated by its BLEU score. Using 14.000 English-Indonesian sentence pairs, the clustering improved the BLEU score of 2.07%.

Information security is an important aspect that must be maintained in order to obtain a successful business. However, financial difficulties and budget constraints create problems in selecting security countermeasures to maintain the network and running the system. Network administrators must rely on labour-intensive processes for tracking network configurations and system vulnerabilities, which needs a lot of expertise and will cause an error prone. Organizational network vulnerabilities and interdependencies are so complex to make traditional vulnerability analysis become not adequate. Decision support capabilities let analysts make the balance between security and availability, and show how best to apply limited security resources. In this paper, we present a methods for strengthening a computer network security.

Problems of scheduling jobs to some machine (Scheduling Jobs on Multiple Machines / SJMM) is one of the classical scheduling problems which can be found in the computing process, especially when done in a distributed computing. Several methods of solving problems has been developed both exact and heuristic approaches (metaheuristik). Tabu Search as one of a relatively new method of metaheuristik can be an alternative method to obtain the settlement approach to these problems. This method has been applied to combinatorial optimization problems, multi ekstermal optimization, and rare event simulation, with results that are optimal solution with a relatively short time. The purpose of this study to develop and implement a Tabu Search method combined with genetic algorithms (Integration Genetic-Tabu Search Algorithm / IGTS) in SJMM problems in computational grid. So that the integration of scheduling algorithms GA and TS can improve processing perpormance Job in grid computing environments. The method used is to include the excess Tabu Search algorithm which formed tabulist to be used in Genetic algorithms. Tabulist used to detect / store data in the process of forming a new population whose job is to detect repeated marriages between same Parent. Results obtained from the algorithm that is designed (IGTS) which serves to determine the allocation of the processing load on the cluster is the increased performance of some value which is quite satisfactory compared with not using tabulist include: makespan = 3.07%, the waiting time = 19.39%, and the number of generations / iterations is smaller.

By using the techniques and methods of Mobile Ad-hoc Network (MANET) each node-client in a wireless network can act as backwarding/forwarding devices; so efficient in the number of nodes connected in the network. The issue has been resolved in this research is the implementation of IPv6 addressing technique based on Dynamic Host Configuration Protocol (DHCP) methods on each node-client in the MANET network. Thus has been proven that the MANET network can be implemented IPv6 and DHCP methods; DHCP itself a technique and method of addressing in a computer network that is currently being developed replace IPv4. The scenario used in this study is to distribute live TV-streaming service from an internet connection to each node-client in the MANET network; and the next following step is measured end-to-end delay parameters on the client side.

This paper describes the decision making process of a robot soccer agent based on the OODA Loop. In this game, an agent that is considered as a player in the field has to decide actions to perform based on its environment. The agent is able to observe the change in the environment using its sensors. Based on this sensory information, a decision for actions has to be done fast and yet accurate. Therefore, the OODA loop method is applied in the decision system to achieve a better architecture of reasoning. The method provide the robosoccer agent abilities to sense the environment, to think, and finally to act.

This paper proposed a interface-based Virtual Tutor Agent (VTA) being able in interaction to multi users in a Problem base learning process in Collaborative virtual environments using MMOG Technology (Massively multiplayer online Game). Here it has been explained concerning two main points associated with conceptual and technology. In the conceptual problem it has been explained a concept associated with Problem based learning, where a learning process has been presented in an interface context in collaborative virtual environments, while in the technology, explained concerning MMOG Technology, among other things the use of multi agent technology, which is focused on the use of Virtual Tutor Agent (VTA) and the applications of AIML (Artificial Intelligence Markup Language), as well as Lua programming language. As for the approach used is by using a simulation method. While the contribution of this Paper is a design and implementation of Problem based learning in collaborative virtual environments in the educational process.

In order to fulfill some tasks to reach a certain common goal, agents need to make sequence of decisions they have to perform as agroup. The decision is taken based on a selection mechanism of available actions. Choosing arbitrary action will lead to time and energy waste, since not all actions are even optimum. Agents need to decide not only which individual action that will lead to optimum performance, but also their joint actions. Applying reinforcement learning in the multiagent's learning process gives a sequence of optimum joint actions, which collaboration of agents based on this optimum joint actions guarantees the fastest time to reach their goal.

Network administrators must rely on labour-intensive processes for tracking network configurations and vulnerabilities, which requires a lot of expertise and error prone. Organizational network vulnerabilities and interdependencies are so complex to make traditional vulnerability analysis become inadequate. Decision support capabilities let analysts make tradeoffs between security and optimum availability, and indicates how best to apply limited security resources. Recent work in network security has focused on the fact that a combination of exploitation is the typical way in which the invader breaks into the network. Researchers have proposed various algorithms to generate graphs based attack tree (or graph). In this paper, we present a framework, Architecture and Approach to Vulnerability Analysis.

LQ45 is a stock market index for Indonesia Stock Exchange (ISX) consists of 45 companies that fulfill certain criteria to aim investor choosing some stocks. However, whether these stocks have tended to rise in price within a specified period. In this paper we analyze some stocks in LQ45 by discrete wavelet transforms. We decomposed signals from closed prices by discrete wavelet transform. Approximation signal from individual stock compared to approximation signal from JCI (Jakarta Composite Index) to justify that it have same trend.

In this research a brief explanation about basic theory of security metrics is given. Implementation of security metrics are also explained. A future work of security metrics fields are given as a guide to other researcher to make an advanced research in the future. A description of security metrics with global taxonomy and classification are discussed.

Communication between the near field communication (NFC) devices occurs in a very close distance of less than 10 cm. In the NFC-based payment system, close proximity between devices will increase the security of transactions. The disadvantage is the interaction between devices requires more physical activity of device owners because the device must be brought near to other devices some times. Besides requiring more physical activity, NFC-based interaction also takes a longer time because the device needs to be moved from one position to another. This paper proposed Secure and efficient protocol that will reduce the physical activity of the device owners and reducing transaction time. The data sending between merchant and payer will be executed without waiting for each other and one transaction will require two data transmissions are performed by the merchant and payer. Transactions are secured by the use of encryption on each data which sent by the merchant and payer. In addition, the protocol also guarantees the security of offline micro transactions and online macros transactions.

The vast amount of information in the Internet is not easy to find and use. Information Extraction technology is one of alternatives that can solve this problem. Conventional Natural Language Processing approach is hampered by its portability, scalability and adaptability. Introduction of Machine Learning into Information Extraction is one of solutions. Inductive Learning only needs annotated training examples. The problem is there is no performance consistency of algorithms on various information domains. Automatic and smart classifier selection from various machine learning algorithms is one of the best way to handle this problem. The goal of this paper is to propose a method for Information Extraction System based on Inductive Learning and Meta Learning that have good performance. In this paper Multi-Inductive Learning is developed to answer that question. Multi-Inductive Learning is consist of several Inductive Learning algorithms that have significant difference in their mechanism. This is to ensure there is bias variance in this method. Through k-fold cross validation on training document, Multi-Inductive Learning algorithm can choose the best classifier for each slot on a certain domain. These best classifiers then employ to do full extraction on testing document. The conducted experiment shows that Multi-Inductive Learning has better performance than that of single Inductive Learning algorithm-based Information Extraction systems. On Reuters Corporate Acquisition, Multi-Inductive Learning gives a score of 46.3 % and has the best performance among other state of the art information systems. Out of nine slots that should be extracted, six of them give the best performance. Multi-Inductive Learning also gives better performance on Job Posting dataset. Average performance of it gives 82.1 % and is the best among other state of the art of Information Extraction. Out of 17 slots that should be tested, nine of them are extracted with the best performance.

Focused crawler is an agent to index information according to specific topic. To traverse WWW, focused crawler makes a prediction of hyperlink's visiting priority in order to download relevant documents as maximum as possible and to minimize downloaded irrelevant documents. Many researchers have proposed methods to improve focused crawling precision by minimizing irrelevant documents. However there is a precision and recall trade-off. More precision the results make less recall. This research has studied on conventional focused crawling search strategy (forward crawling) and Web documents structure. The result shows the low recall of conventional focused crawling is caused by some structural characteristics of WWW. Therefore, this research proposes a new strategy of focused crawler. The new strategy is a combination of bidirectional (forward and backward) crawling and bibliometric concepts (co-citation & co-reference). Bidirectional crawling is to improve the exploration and co-citation & co-reference concepts are to control the focusing. With this new strategy, focused crawler can obtain relevant documents that are connected through co-citations or relevant communities that act connected through co-references. Based on experiments that have been carried out, the results show that focused crawler with this new strategy, named CT-FC (more Comprehensive Traversal Focused Crawler) has better exploration capability so that recall increases and precision can remain high.

Part of speech (PoS) is the classification of words according to form, function, and meaning. POS tagging is an important problem and is the first step in natural language processing. POS tagging directly affect the accuracy on the next steps in the processing of natural language like syntax parsing, ambiguity of the word, and machine translation

One of early processes in an embedded system designing is system modeling. A modeling is performed to get preliminary figure of a product before it could be implemented in form of a prototype. Generally speaking, an embed system designing processes comprise five stages: product specifications, HW/SW separation. detailed HW/SW designing, integration and testing. The final output is in form of detailed RTL. After the RTL has been completed, the improvement processes could be started. This paper proposes a new designing process by performing UML-based preliminary designing and TLM preliminary process before detailed HW/SW designing. The correction processes would be performed only under the TML Model. It is expected that by this method we could perform the overall designing processes faster than the conventional method. We use SystemC as a modeling Language.

MAM (multi-mode access memory) is a memory that has the characteristics of PSMU (parallel search and multiple update). This new type of memory is applied to LAPCAM (linear array of processors with content addressable memory) architecture for image processing. Using MAM, the global and regional type of image processing algorithms can be developed efficiently. MAM consists of 3 different types of memory i.e. conventional memory (RAM), content addressable memory (CAM), and shift/rotate memory (FIFO). The objectives of the research are focused on the design of MAM memory, implementation of the design in VHDL (VHSIC hardware description language), verification of the design by simulation using software active-HDL and evaluation of the simulation results.