Research Seminar Series: I Feel You: State-of-the-art in Emotion Modelling
1000 until 1100
Meeting Room 7th Floor
Dr. Syaheerah Lebai Lutfi
It is easy to get frustrated at machines, perhaps because they seem to be callous. By and large, the quality of human-computer interaction is affected due to the inability of the computerized agents to recognise and adapt to user emotional state (or sometimes, even personality). Now with the mass appeal of artificially-mediated communication, there has been an increasing need for machines to be socially and emotionally intelligent, that is, to infer
and adapt to their human interlocutors' emotions on the fly, in order to ascertain an affective, empathetic and naturalistic interaction. Recent studies has shown synthetic agents (be it a robot or a virtual agent) that are affect-sensitive have offer great advantages, from reducing user frustration to increasing learner’s engagement. These reasons have motivated the development of artificial agents towards including socio-emotional elements, turning them into affective and socially-sensitive interfaces.
This talk will discuss the state-of-art in Affective Computing including the effort towards culture-sensitive affective interfaces. To facililate listeners’ understanding, a sample of a recent study will be presented.
For more information, please visit www.syaheerah.com
Proposal Review: Optimizing Crowd Evacuation Plan in the Emergency Route Planning Problems
1000 until 1130
Viva Room 7th Floor
Mohd Nor Akmal bin Khalid
Dr. Umi Kalsom Yusof
Disastrous situations, either occurred naturally (such as fires, earthquake, floods, hurricane) or man-made (such as terrorist bombings, chemical spills, etc.), have claimed the lives of thousands in the last millennium. Emergency evacuation is an essential strategy in managing these disastrous situations. From top-down perspective of emergency evacuation, several stages are involves which aggregated as strategic, tactical, and operational. In addition, to mitigate risks and reduce potential fatalities, the operational perspectives of emergency evacuation involves operations such as mitigation, preparedness, response, and recovery. Typically, factors involved in the emergency evacuation include the operational condition, routing complexity, crowd characteristics, environments, and the disaster itself. As such, optimizing the evacuation operations during emergency situation would require an effective crowd evacuation plan, which is acknowledged as one of the vital study of the societal research as well as emergency route planning (ERP) community. This study attempts to contribute to the body of knowledge and the general community by optimizing the ERP problems, supporting the betterment of evacuation routine while reducing the cost of life. Exhaustive prior works have been studied and presented, while the principal research aims had also been made. Various classifications of priory developed approaches for emergency evacuation that encompassed the needs of variety of public community as well as fulfilling the complexity of the situation, are summarized and discussed. Three components of interest is concentrated, which are demand (crowd), resource (roadways/networks), and event (disaster). The direction and trend of the work had been identified which leads to the realization of the potential crowd evacuation approach. The methodology used for the study had also been discussed and elaborated in details. Lastly, the expected deliverable are discussed for the upcoming stage of this work.
Research Review: Enhanced Intelligent Water Drops Algorithms and Their Applications to Optimization Problems
1500 until 1630
Meeting Room 7th Floor
Basem O. F. Alijla
Prof. Dr. Ahamad Tajudin Khader
Prof. Dr. Lim Chee Peng, Dr. Wong Li Pei
The Intelligent Water Drop (IWD) algorithm is a recent natural-inspired stochastic swarm-based model that is useful for undertaking optimization problems. It imitates some of the natural phenomena of water flow in a river bed. Since its inception, the IWD algorithm has been successfully tailored to solve several discrete and continuous optimization problems. It is characterized by using a cooperative learning model to construct better solutions over consecutive generations and address exploration and exploitation issues pertaining to the search space. The main aim of this research is to enhance the IWD algorithm and overcome its limitations pertaining to local optimal, premature convergence, population diversity, as well as balanced exploration and exploitation in handling optimization problems. Firstly, instead of using the fitness proportionate selection method in the solution construction phase of the original IWD algorithm, two ranking-based selection methods, i.e. linear ranking and exponential ranking, are proposed in this research. Both ranking-based selection methods of the modified-IWD algorithm aim to solve the identified limitations of the fitness proportionate selection method, and enable the original IWD algorithm to escape from local optima and avoid premature convergence. Secondly, a Master River and Multiple Creeks (MRMC) model is proposed for the modified-IWD algorithm. The resulting model, abbreviated as MRMC-IWD, comprises one master river (swarm) that cooperates with several independent creeks (sub-swarms) in an attempt to maintain a balanced exploitation and exploration search process for solving optimization problems. Each creek provides a partial solution to the master river in a sequential manner. The main benefits of MRMC-IWD include enhancing population diversity and preserving a balanced exploration and exploitation search process of the modified IWD algorithm. In addition, MRMC-IWD is further enhanced with a local search algorithm, and the resulting model is called Hybrid-MRMC-IWD. To evaluate the usefulness of the enhanced IWD-based models, three combinatorial optimization problems, i.e., rough set feature subset selection (RSFS), MKP, and TSP problems, are considered. A series of experiments is conducted, and the results demonstrate that the proposed models i.e. modified-IWD, MRMC-IWD, and Hybrid-MRMC-IWD, are able to avoid local optimal and premature convergence as well as to enhance population diversity and preserve a balanced exploitation and exploration search process; therefore improving the performance of the original IWD algorithm.
Research Review: Enhancing the Quality of Service in Moving Networks based on NEMO Basic Support Protocol
1500 until 1630
Viva Room 7th Floor
Badiea Abdulkarem Mohammed Al-Shaibani
Assoc. Prof. Dr. Wan Tat Chee
To fulfil the need for on-the-move and uninterrupted internet connectivity in Mobile Networks instead of the end-host mobility, IETF NEMO working group was created to extend basic end-host mobility support in Mobile IPv6 (MIPv6). NEMO Basic Support Protocol (NEMO) has been standardized by this group to provide the network mobility support. This protocol brings forward connectivity to all the mobile nodes within a Mobile Router (MR). When MR changes its connection point to a new point the handover procedure has to be conducted to maintain the link switching and to inform MR’s Home Agent (HA) in the Home Network by the new Care-of-Address (CoA). However, the handover latency in NEMO is high because this protocol focuses in network-layer (L3) handover phase and this phase starts after the link-layer (L2) handover phase is completed.
NEMO Basic Support Solution would be supposed to support transparent mobility to mobile network nodes (MNNs) in mobile networks by using MR-HA bi-directional tunnelling. However, if multiple mobile networks are nested, that brings a routing overhead to network which is well known as "pinball" routing problem. The nested tunnels’ problem in the nested NEMO networks is not considered in the main specification of this protocol.
Many schemes have been proposed to solve these problems by optimizing the handover signaling procedure, and by proposing routing optimization scheme for NEMO. This research propose a new framework the combine better optimized signaling handover procedure, and a proposed Routing Optimization scheme as a solution for the lack of the nested tunnels’ problem.
Analytical results highlight the importance of the proposed schemes comparing to others are provided, revealing that the proposed scheme has enhanced the handover latency and the disruption time. Simulated experiments using OMNet++ 4.2.2 simulation tools have been carried out to evaluate the proposed framework. The results from the experiments showed that the handover latency, the disruption time, the packets delay, and the packets lost in the proposed frame work have been enhanced comparing to other schemes.