MERCon - 2016

Permanent URI for this collectionhttp://192.248.9.226/handle/123/14697

Browse

Now showing 1 - 20 of 75

item: Conference-Full-text
2016 Moratuwa Engineering Research Conference (MERCon) [Pre Text}
(IEEE, 2016)
item: Conference-Full-text
Categorizing food names in restaurant reviews
(2016-04) Prakhash, S; Nazick, A; Panchendrarajan, R; Brunthavan, M; Ranathunga, S; Pemasiri, A; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
There are many aspects such as food, service, and ambience that a customer would look for, when deciding on a restaurant to dine in. Among these aspects, the type of food it sells and the food quality are the most important. Therefore, when automatically rating restaurants based on customer reviews, the food aspect plays a major role. There exists some research on rating individual food items in a restaurant. However, a potential customer requires not the ranking of an individual food item, but the ranking of a particular food category in general. In order to do that, a categorization of food names is required. This paper presents two techniques for food name categorization using document similarity measurements.
item: Conference-Full-text
Machine learning approach to recognize subject based sentiment values of reviews
(IEEE, 2016-04) De Mel, NM; Hettiarachchi, HH; Madusanka, WPD; Malaka, GL; Perera, AS; Kohomban, U; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Due to the increase in the number of people participating online on reviewing travel related entities such as hotels, cities and attractions, there is a rich corpus of textual information available online. However, to make a decision on a certain entity, one has to read many such reviews manually, which is inconvenient. To make sense of the reviews, the essential first step is to understand the semantics that lie therein. This paper discusses a system that uses machine learning based classifiers to label the entities found in text into semantic concepts defined in an ontology. A subject classifier with a precision of 0.785 and a sentiment classifier with a correlation coefficient of 0.9423 was developed providing sufficient accuracy for subject categorization and sentiment evaluation in the proposed system.
item: Conference-Full-text
An automatic classifier for exam questions with wordnet and cosine similarity
(IEEE, 2016-04) Jayakodi, K; Bandara, M; Meedeniya, D; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
The learning objectives, learning activities and assessment are very much interrelated. Assessment helps to evaluate students learning achievement. Poorly designed assessments usually fail to examine the achievement of intended learning outcome of a course. There are different taxonomies that have been developed to identify the level of the assessment being practiced such as Bloom’s and SOLO. In this research we have studied the use of WordNet with Cosine similarity algorithm for classifying a given exam question according to Bloom’s taxonomy learning levels. WordNet similarity algorithm depends on the extracted verbs from exam question. Cosine similarity algorithm was based on identification of question patterns of exam question. It consists of tag pattern generation module, grammar generation module, parser generation and cosine similarity checking module. This algorithm was helpful to classify the exam question where verbs were not present in exam questions. Exam questions taken from courses at the Department of Computing and Information Systems at Wayamba University were used as a basis for a performance comparison, with the autonomous system providing classifications that were consistent with those provided by domain experts on approximately 71% of occasions.
item: Conference-Full-text
Support for traceability management of software artefacts using natural language processing
(IEEE, 2016-04) Arunthavanathan, A; Shanmugathasan, S; Ratnavel, S; Thiyagarajah, V; Perera, I; Meedeniya, D; Balasubramaniam, D; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
One of the major problems in software development process is managing software artefacts. While software evolves, inconsistencies between the artefacts do evolve as well. To resolve the inconsistencies in change management, a tool named “Software Artefacts Traceability Analyzer (SATAnalyzer)” was introduced as the previous work of this research. Changes in software artefacts in requirement specification, Unified Modelling Language (UML) diagrams and source codes can be tracked with the help of Natural Language Processing (NLP) by creating a structured format of those documents. Therefore, in this research we aim at adding an NLP support as an extension to SAT-Analyzer. Enhancing the traceability links created in the SAT-analyzer tool is another focus due to artefact inconsistencies. This paper includes the research methodology and relevant research carried out in applying NLP for improved traceability management. Tool evaluation with multiple scenarios resulted in average Precision 72.22%, Recall 88.89% and F1 measure of 78.89% suggesting high accuracy for the domain.
item: Conference-Full-text
Cheap food or friendly staff? weighting hierarchical aspects in the restaurant domain
(IEEE, 2016-05) Panchendrarajan, R; Murugaiah, B; Prakhash, S; Ahamed, MNN; Ranathunga, S; Pemasiri, A; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
In aspect-level opinion mining, each aspect is assigned a rating based on customer reviews. More often than not, these aspects exhibit a hierarchical relationship, and the restaurant domain is no difference. With the existence of such hierarchical relationships, rating of an aspect is based on the composite score of its sub-elements. However, the influence of these sub-aspects on the score of a parent aspect is not uniform, since some sub-aspects are perceived more important than others. Therefore, when calculating the composite score for an aspect, influence of each sub-aspect should be weighted according to its perceived importance. Identifying weights for different aspects is addressed as the problem of multi-attribute weighting. However the existing approaches do not utilize the relationships between aspects to find weights. This paper presents an approach to find weights for aspects that exhibit hierarchical relationships in restaurant domain using an improved version of the Analytic Hierarchy Process (AHP), one of the Multi Attribute Decision Making Techniques (MADTs). Different aspects of the restaurant domain are modeled as a hierarchy and weights for aspects are calculated using AHP. Occurrence counts of aspects in restaurant reviews are used to obtain the relative importance of aspects. This approach provides acceptable consistency ratios for the pairwise comparison matrices obtained for each level in the hierarchy of aspects.
item: Conference-Full-text
Ananya - a named-entity-recognition (ner) system for sinhala language
(IEEE, 2016-04) Manamini, SAPM; Ahamed, AF; Rajapakshe, RAEC; Reemal, GHA; Jayasena, S; Dias, GV; Ranathunga, S; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Named-Entity-Recognition (NER) is one of the major tasks under Natural Language Processing, which is widely used in the fields of Computer Science and Computational Linguistics. However, the amount of prior research done on NER for Sinhala is very minimal. In this paper, we present data-driven techniques to detect Named Entities in Sinhala text, with the use of Conditional Random Fields (CRF) and Maximum Entropy (ME) statistical modeling methods. Results obtained from experiments indicate that CRF, which provided the highest accuracy for the same task for other languages outperforms ME in Sinhala NER as well. Furthermore, we identify different linguistic features such as orthographic word level and contextual information that are effective with both CRF and ME Algorithms.
item: Conference-Full-text
New mems based micro gripper using sma for micro level object manipulation and assembling
(IEEE, 2016-04) Munasinghe, KC; Bowatta, BGCT; Abayarathne, HYR; Kumararathna, N; Maduwantha, LKAH; Arachchige, NMP; Amarasinghe, YWR; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Micro electro mechanical systems (MEMS) based micro gripper for micro-assembling processes is designed. Gripper arms are actuated using a combination of Ni-Ti shape memory alloys and Si springs allowing a gripping range of 1- 120μm. Heat which is needed to actuate shape memory alloy (SMA) is obtained by electro thermal properties of Ni-Ti by passing a current which is controlled by specially designed microprocessor controlled circuit design. Stability of the gripper design is validated using a commercially available finite element analysis (FEA) tool COMSOL Multiphysics simulations for stresses in gripper arms and for Eigen frequencies. Proposed fabrication process uses minimum number of masks and it makes the design much suits for batch production.
item: Conference-Full-text
Design and simulation of mems based 5-dof tactile force sensor
(IEEE, 2016-04) Udayanga, TDI; Jayathilaka, WADM; Amarasinghe, YWR; Dao, DV; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
This paper describes design and simulation of five degrees of freedom (5-DOF) Micro-Electro-Mechanical systems (MEMS) based tactile force sensor. Tactile sensing involves with measuring physical parameters such as force, temperature, etc. with the aid of physical touch. Over the past decades tactile sensors are gaining popularity over non-contact sensors in biomedical and robotic applications. Proposed sensor design with 3mm x 3mm x 300μm dimensions, has the capability to measure not only the magnitude but also the direction of the force applied. A wagon wheel spring structure was proposed, where 8 beams work as springs to relief the force applied. Behavior of these 8 beams are monitored under each loading conditions using defused piezoresistive sensing elements. A finite element analysis of structure was performed to optimize and validate the structure and Multiphysics analysis was performed to validate the working principal of the proposed sensor.
item: Conference-Full-text
A novel mems motor based on thermal actuation
(IEEE, 2016-04) De Silva, AHTE; De Silva, DDN; Perera, KDCJ; Priyashantha, AMB; Sampath, LLR; Darshan, P; Jayathilaka, WADM; Amarasinghe, YWR; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
It is essential for the Micro Electro Mechanical Systems (MEMS) research industry to introduce novel concepts of micro motors to overcome problems in existing micro motors. This paper propose a novel concept of a micro motor using kink actuators and reciprocating rack and pinion assembly. Design details of the reciprocating rack and pinion is discussed in the paper. A detailed analysis of structural, transient, thermal, electrical properties was performed using the COMSOL software is also discussed. Finally a fabrication method is purposed using electron beam lithography and ultraviolet lithography.
item: Conference-Full-text
Smart solar tracking and on-site photovoltic efficiency measurement system
(IEEE, 2016-04) Basnayake, BADJCK; Jayathilaka, WADM; Amarasinghe, YWR; Attalage, RA; Jayasekara, AGBP; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
On-site photovoltaic efficiency data is a valuable asset during a process of predicting photovoltaic potential. Not just the solar power output, but also the ambient conditions and panel temperature should be measured for a better and convinced results. Due to the unavailability of on-site data, erroneous conclusions have been made after various prediction methods. Smart solar tracking and on-site photovoltaic measurement system is proposed as a novel tool to be used in solar potential predictions which can measure and log on-site solar data. This device is capable of measuring and logging available solar power together with ambient measurements such as light intensity level, ambient temperature and humidity level and panel temperature. Measured data will then be stored in an internal memory card and will be available at any moment. Integrated wireless communication module will enable remote log-in and control of the device. Computer based Graphical User Interface (GUI) software application enables the remote access to the gathered data and optimization of its operation.
item: Conference-Full-text
Development of a numerically controlled hot wire foam cutting machine for wing mould construction
(IEEE, 2016-04) Abeysinghe, A; Abeysiriwardena, S; Nanayakkarawasam, R; Wimalsiri, W; Lalitharatne, TD; Tennakoon, S; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
The foam mould making for Unmanned Aerial Vehicles should be efficient and accurate to improve research and manufacturing. Hot wire cutting is a widely used method in foam cutting. In hot wire cutting, accuracy and quality of the foam cut mainly depends on the variable cutting parameters which affect the cutting process. To perform a proper cut with required quality the cutting parameters should be set precisely and accurately. This research is done to identify the variations and inter dependency of cutting parameters. It is important to estimate appropriate cutting parameter values before the actual cut. The results of this research aid to improve the hot wire foam cutting by solving the limitations and drawbacks to select best cutting parameters of the Computer Numerically Controlled machine which is in progress.
item: Conference-Full-text
Model-based input-adaptive vectorization
(2016-04) Sundararajah, K; Jayasena, S; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
In a program, not all the bits of a variable are always used during execution. Identifying the minimum number of bits necessary to represent a variable in a program can potentially provide optimization opportunities. Providing the knowledge of bitwidths to a compilation and execution framework will be advantageous if it could use that information to optimize the execution of the program, for instance, being able to select instructions for SIMD vectorization. This paper introduces a framework to exploit the potential vectorizations hidden in a program which is not exposed during static compilation time. Our framework unlocks instruction level data parallelism by using the bitwidths of array like variables that depend on runtime input. Our framework shows a maximum achievable performance gain of 37% and a mean achievable performance gain of 11% against the ICC compiler on our micro benchmark suite.
item: Conference-Full-text
A visualization and analysis platform for performance tuning
(IEEE, 2016-04) Eranjith, HMD; Fernando, ID; Fernando, GKS; Soysa, WCM; Jayasena, VSD; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
With a framework like OpenTuner, one could build domain-specific multi-objective program auto-tuners and gain significant performance improvements. But explaining why and interpreting the results are often hard, mainly due to the large number of parameters and the inability to figure out how each parameter affects the performance improvement. We have a solution that can explain the performance improvements by identifying key parameters while providing better insights on the tuning process. Our tool uses machine learning techniques to identify parameters which account for a significant performance improvement. A user could utilize different methods provided in the tool to further experiment and verify the accuracy of such findings. Further, our tool uses multidimensional scaling to display all the configurations in a two dimensional graph. This interface allows users to analyze the search space closely and identify clusters of configurations with good or bad performance. It also provides real-time information of tuning process which would help users to optimize the tuning process.
item: Conference-Full-text
Reducing computational time of closed-loop weather monitoring: a complex event processing and machine learning based approach
(IEEE, 2016-04) Chandrathilake, HMC; Hewawitharana, HTS; Jayawardana, RS; Viduranga, ADD; Bandara, HMND; Marru, S; Perera, S; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Modern weather forecasting models are developed to maximize the accuracy of forecasts by running computationally intensive algorithms with vast volumes of data. Consequently, algorithms take a long time to execute, and it may adversely affect the timeliness of forecast. One solution to this problem is to run the complex weather forecasting models only on the potentially hazardous events, which are pre-identified by a lightweight data filtering algorithm. We propose a Complex Event Processing (CEP) and Machine Learning (ML) based weather monitoring framework using open source resources that can be extended and customized according to the users’ requirements. The CEP engine continuously filters out the input weather data stream to identify potentially hazardous weather events, and then generates a rough boundary enclosing all the data points within the Areas of Interest (AOI). Filtered data points are then fed to the machine learner, where the rough boundary gets more refined by clustering it into a set of AOIs. Each cluster is then concurrently processed by complex weather algorithms of the WRF model. This reduces the computational time by ~75%, as resource heavy weather algorithms are executed using a small subset of data that corresponds to only the areas with potentially hazardous weather.
item: Conference-Full-text
Proaasel: prospect theory based continuous authentication attribute selection model
(IEEE, 2016-04) Premarathne, US; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Existing continuous authentication models use a fixed set of attributes and do not consider the application specific requirements and associated vulnerabilities in their selection. Selecting appropriate attributes for continuous authentication is essentially a multi-criteria decision making process. Existing multi-criteria decision making models are less competent in providing a preference for each attribute in a set of possible attributes. In this paper we propose a utility based approach: PROAASEL, prospect theory based continuous authentication attribute selection model. The main assumption of our approach is the associated risks for each attribute are pre-defined in terms of known vulnerabilities. The main advantage of our model is the ability to select the attributes based on application specific risk characterizations. We have evaluated PROAASEL using CVE data from [1]. Furthermore, we compared the selection method with existing MCDM techniques TOPSIS and N-model for plausible application scenarios. The results reveal that PROAASEL is more expressive and offer more reliable selection when the associated risks are fixed.
item: Conference-Full-text
Fuel consumption prediction of fleet vehicles using machine leaning: a comparative study
(IEEE, 2016-04) Wickramanayake, S; Bandara, HMND; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Ability to model and predict the fuel consumption is vital in enhancing fuel economy of vehicles and preventing fraudulent activities in fleet management. Fuel consumption of a vehicle depends on several internal factors such as distance, load, vehicle characteristics, and driver behavior, as well as external factors such as road conditions, traffic, and weather. However, not all these factors may be measured or available for the fuel consumption analysis. We consider a case where only a subset of the aforementioned factors is available as a multi-variate time series from a long distance, public bus. Hence, the challenge is to model and/or predict the fuel consumption only with the available data, while still indirectly capturing as much as influences from other internal and external factors. Machine Learning (ML) is suitable in such analysis, as the model can be developed by learning the patterns in data. In this paper, we compare the predictive ability of three ML techniques in predicting the fuel consumption of the bus, given all available parameters as a time series. Based on the analysis, it can be concluded that the random forest technique produces a more accurate prediction compared to both the gradient boosting and neural networks.
item: Conference-Full-text
Applying agile practices to avoid chaos in user acceptance testing: a case study
(IEEE, 2016-04) Padmini, KVJ; Perera, I; Bandara, HMND; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
Agile practices have gained increasing popularity in Information Technology (IT), Education, Marketing, and Advertising industry, as it brings quality products into the market faster. Scrum, Lean Development, and Extreme Programming are the most commonly considered processes under the Agile umbrella. Scrum or scrum variants form a high performance, collaborative team to handle projects that are more complex. We examine the applicability of the scrum framework to a large-scale revenue management system for User Acceptance Testing (UAT). Industry believes integration and acceptance testing is not easy to perform within the scrum framework. Nevertheless, very little is explored about the acceptance testing in Agile practices. We fill this gap by empirically evaluating UAT of a complex, large-scale system (in a public sector organization) to showcase the applicability of scrum framework. While the initial UAT team consisted of 100 domain experts, no process was defined for the UAT. This made it easier to streamline the UAT into the scrum framework. Once the scrum framework was introduced significant improvements in the UAT team was experienced with improved morale, productivity, efficiency, and time to market while having a smooth flow.
item: Conference-Full-text
Open source mobile network
(IEEE, 2016-04) Sajeev, U; Muralitharan, R; Ramsan, MMM; Zamrath, MZM; Dias, D; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
We present the development and testing of a cellular network built on open source technologies. This network supports voice calls and text messages within the network or between networks. The main objective of this work is to develop a stand-alone mobile network consisting of one base station and a switch, the access to which can be controlled as required. The mobile network is also connected to public networks. The designed network is based on Global System for Mobile Communications (GSM). Our work includes the configuration of the embedded hardware components, selection and integration of the proper software combination, testing of compatible versions of the software, hardware-software integration within the embedded system, and interconnecting the systems via the Internet. The key contribution and novelty of the project is the development of a stand-alone mobile network base station/ switch using the Universal Software Radio Peripheral (USRP) and a Banana Pi device.
item: Conference-Full-text
Improved ldpc decoding algorithms based on min-sum algorithm
(IEEE, 2016-04) Kumara, YVAC; Wavegedara, CB
Low-Density Parity check (LDPC) codes offer highperformance error correction near the Shannon limit which employs large code lengths and some iterations in the decoding process. The conventional decoding algorithm of LDPC is the Log Likelihood Ratio based Belief Propagation (LLR BP) which is also known as the 'Sum-Product algorithm' which gives the best decoding performance and requires the most computational complexity and implementations with increased hardware complexity. Another simpler variant of this algorithm is used which is known as 'min-sum algorithm' which reduces computational complexity as well as hardware complexity but with reduced accuracy. This paper analyzes the reason min-sum algorithm is more prone to errors when compared to the sumproduct algorithm, and puts forward two improved algorithms which improve the performance of the min-sum algorithm with comparable algorithmic complexity.

Browse

Recent Submissions