• Techniques for improving the labelling process of sentiment analysis in the Saudi stock market

      AL-Rubaiee, Hamed Saad; Qiu, Renxi; Alomar, Khalid; Li, Dayou; University of Bedfordshire; King Abdulaziz University (Science and Information Organization, 2018-12-31)
      Sentiment analysis is utilised to assess users' feedback and comments. Recently, researchers have shown an increased interest in this topic due to the spread and expansion of social networks. Users' feedback and comments are written in unstructured formats, usually with informal language, which presents challenges for sentiment analysis. For the Arabic language, further challenges exist due to the complexity of the language and no sentiment lexicon is available. Therefore, labelling carried out by hand can lead to mislabelling and misclassification. Consequently, inaccurate classification creates the need to construct a relabelling process for Arabic documents to remove noise in labelling. The aim of this study is to improve the labelling process of the sentiment analysis. Two approaches were utilised. First, a neutral class was added to create a framework of reliable Twitter tweets with positive, negative, or neutral sentiments. The second approach was improving the labelling process by relabelling. In this study, the relabelling process applied to only seven random features (positive or negative): "earnings" (Arabic source), "losses" (Arabic source), "green colour" (Arabic source:Arabic source), "growing" (Arabic source), "distribution" (Arabic source), "decrease" (Arabic source), "financial penalty" (Arabic source), and "delay" (Arabic source). Of the 48 tweets documented and examined, 20 tweets were relabelled and the classification error was reduced by 1.34%.
    • Templated assembly of micropatterned Au-Ni nanoparticles on laser interference-structured surfaces by thermal dewetting

      Wang, Lu; Dong, Litong; Li, Li; Ding, Ran; Liu, Jinyun; Zhang, Wenxiao; Wang, Ying; Weng, Zhankun; Guo, Xudong; Wang, Zuobin; et al. (Elsevier, 2019-09-18)
      This paper introduces a laser-interference-controlled electrochemical deposition method for direct fabrication of periodically micropatterned magnetite (Fe3O4) nanoparticles (NPs). In this work, Fe3O4 NPs were controllably synthesized on the areas where the photoconductive electrode was exposed to the periodically patterned interferometric laser irradiation during the electrodeposition. Thus, the micropattern of Fe3O4 NPs was controlled by interferometric laser pattern, and the crystallization of the particles was controlled by laser interference intensity and electrochemical deposition conditions. The bottom-up electro- chemical approach was combined with a top-down laser interference method- ology. This maskless method allows for in situ fabrication of periodically patterned magnetite NPs on the microscale by electrodeposition under room temperature and atmospheric pressure conditions. In the experiment, Fe3O4 NPs with the mean grain size below 100 nm in the pattern of 5-lm line array were achieved within the deposition time of 100 s. The experiment results have shown that the proposed method is a one-step approach in fabricating large areas of periodically micropatterned magnetite NPs. 
    • Text-independent speaker identification using vowel formants

      Almaadeed, Noor; Aggoun, Amar; Amira, Abbes (Springer New York LLC, 2015-05-05)
      Automatic speaker identification has become a challenging research problem due to its wide variety of applications. Neural networks and audio-visual identification systems can be very powerful, but they have limitations related to the number of speakers. The performance drops gradually as more and more users are registered with the system. This paper proposes a scalable algorithm for real-time text-independent speaker identification based on vowel recognition. Vowel formants are unique across different speakers and reflect the vocal tract information of a particular speaker. The contribution of this paper is the design of a scalable system based on vowel formant filters and a scoring scheme for classification of an unseen instance. Mel-Frequency Cepstral Coefficients (MFCC) and Linear Predictive Coding (LPC) have both been analysed for comparison to extract vowel formants by windowing the given signal. All formants are filtered by known formant frequencies to separate the vowel formants for further processing. The formant frequencies of each speaker are collected during the training phase. A test signal is also processed in the same way to find vowel formants and compare them with the saved vowel formants to identify the speaker for the current signal. A score-based scheme allows the speaker with the highest matching formants to own the current signal. This model requires less than 100 bytes of data to be saved for each speaker to be identified, and can identify the speaker within a second. Tests conducted on multiple databases show that this score-based scheme outperforms the back propagation neural network and Gaussian mixture models. Usually, the longer the speech files, the more significant were the improvements in accuracy.
    • Theory-based user modeling for personalized interactive information retrieval

      Ullah, Asad; Liu, Haiming; University of Bedfordshire (2016-07-17)
      In an effort to improve users’ search experiences during their information seeking process, providing a personalized information retrieval system is proposed to be one of the effective approaches. To personalize the search systems requires a good understanding of the users. User modeling has been approved to be a good method for learning and representing users. Therefore many user modeling studies have been carried out and some user models have been developed. The majority of the user modeling studies applies inductive approach, and only small number of studies employs deductive approach. In this paper, an EISE (Extended Information goal, Search strategy and Evaluation threshold) user model is proposed, which uses the deductive approach based on psychology theories and an existing user model. Ten users’ interactive search log obtained from the real search engine is applied to validate the proposed user model. The preliminary validation results show that the EISE model can be applied to identify different types of users. The search preferences of the different user types can be applied to inform interactive search system design and development.
    • Threshold power based UE admittance and contention free resource allocation for interference mitigation in cognitive femtocells

      Safdar, Ghazanfar Ali; Tariq, Faisal; Kpojime, Harold Orduen; University of Bedfordshire; Queen Mary University of London (Springer, 2017-11-17)
      Femtocells are aimed at providing strong coverage in the indoor area where typical macrocell coverage is very poor. It has hugely attracted network operators and stakeholders mainly due to its simple plug and play operation and low cost. Femtocells operate on a much lower power compared to macrocell and thus provide a number of benefits including energy efficiency and frequent spatial reuse of the spectrum. Femtocells are overlaid on macrocells and designed to co-exist with them sharing the same spectrum pool. However, since they are deployed by the end user, no pre-deployment resource planning is possible. So, interference among the femtocells as well as between femtocells and macrocells remain a major bottleneck for successful operation of femtocell networks. This paper proposes a novel threshold power based admittance and contention free resource allocation for interference mitigation in cognitive femtocell networks. In our proposed scheme, a Femtocell Access Point with Cognitive radio capability known as Cognitive Femtocells (CF), sets a threshold value on the mutual interference between itself and a close by macrocell user equipment (MUE). To mitigate cross-tier interference, a CF classifies MUEs with higher than threshold interference value as Undesirable MUEs (UMUEs) and subsequently admits it as one of its user equipment. MUEs with lower than threshold interference values are classified as Desirable MUEs (DMUEs). To mitigate co-tier interference, proposed scheme introduces a scheduling engine which employs matching policy attributes and assigns resource blocks (RBs) of unique DMUEs to CFs to avoid any possible contention problems, thus providing improved co-tier interference. System level simulations have been performed to demonstrate effectiveness of scheme and significant performance improvement in terms of SINR, throughput and spectrum efficiency.
    • Time series chlorophyll-A concentration data analysis: a novel forecasting model for aquaculture industry

      Eze, Elias Chinedum; Kirby, Sam; Attridge, John; Ajmal, Tahmina; University of Bedfordshire; Chelsea Technology Group (2021-06-29)
      Eutrophication in fresh water has become a critical challenge worldwide and chlorophyll-a content is a key water quality parameter that indicates the extent of eutrophication and algae concentration in a body of water. In this paper, a forecasting model for the high accuracy prediction of chlorophyll-a content is proposed to enable aquafarm managers to take remediation actions against the occurrence of toxic algal blooms in the aquaculture industry. The proposed model combines the ensemble empirical mode decomposition (EEMD) technique and a deep learning (DL) long short-term memory (LSTM) neural network (NN). With this hybrid approach, the time-series data are firstly decomposed with the aid of the EEMD algorithm into manifold intrinsic mode functions (IMFs). Secondly, a multi-attribute selection process is employed to select the group of IMFs with strong correlations with the measured real chlorophyll-a dataset and integrate them as inputs for the DL LSTM NN. The model is built on water quality sensor data collected from the Loch Duart salmon aquafarm in Scotland. The performance of the proposed novel hybrid predictive model is validated by comparing the results against the dataset. To measure the overall accuracy of the proposed novel hybrid predictive model, the Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) were used.
    • Timeline and episode-structured clinical data: pre-processing for Data Mining and analytics

      Lu, Jing; Hales, Alan; Rew, David; Keech, Malcolm; Southampton Solent University; University Hospital Southampton; University of Bedfordshire (Institute of Electrical and Electronics Engineers Inc., 2016-06-23)
      Data Mining has been used in the healthcare domain for diagnosis and treatment analysis, resource management and fraud detection. It brings a set of tools and techniques that can be applied to large-scale patient data to discover underlying patterns and provide healthcare professionals an additional source of knowledge for making decisions. The Southampton Breast Cancer Data System (SBCDS) containing some 16,000 timeline-structured records is a visually rich and highly intuitive system for the manual and automated transfer of demographic, pathology and treatment data into an episode-based structure. While expansion of the data mining capability in SBCDS is one of the objectives of our research, real-world patient data is generally incomplete, inconsistent and containing errors. This case study will focus on the data pre-processing stage in order to clean the raw data and prepare the final dataset for use in data mining and analytics. Some initial results are given for sequential patterns mining and classification which highlight the advantages of the approach.
    • Timely and efficient multihop broadcast scheme for reliable inter-vehicular communication

      Muhammad, Shehu; Zhang, Sijing; Dyo, Vladimir; University of Bedfordshire (Institute of Electrical and Electronics Engineers Inc., 2019-07-04)
      Majority of vehicular safety applications such as collision avoidance and blind spot warnings require a limited delay for a message to be transmitted from the abnormal vehicle to the endangered vehicles to avert the occurrence of the accident. This paper proposed a Timely and Efficient Multihop Broadcast (TEMB) Scheme, which combines an adaptive forwarding strategy that depends on the estimated transmission range between moving vehicles on the road. The scheme further combines ARQ with an adaptive Forward Error Correction system to minimise the number of retransmissions while ensuring the reliability of message delivery to the intended recipient. Modifications on the IEEE802.11p data and acknowledgement (feedback) frames are presented. The scheme has been evaluated and compared with the Fast Broadcast scheme, the results have shown that the TEMB scheme has outperformed the Fast Broadcast scheme for reliable and timely message delivery among the moving vehicles on the motorway.
    • Timely and reliable packets delivery over Internet of Vehicles (IoVs) for road accidents prevention: a cross-layer approach

      Eze, Elias Chinedum; Zhang, Sijing; Liu, Enjie; Nweso, Emmanuel N.; Eze, Joy C.; University of Bedfordshire; Ebonyi State University, Nigeria (IET, 2016-09-01)
      With the envisioned era of Internet of Things (IoTs), all aspects of Intelligent Transportation Systems (ITS) will be connected to improve transport safety, relieve traffic congestion, reduce air pollution, enhance the comfort of transportation and significantly reduce road accidents. In IoVs, regular exchange of current position, direction, velocity, etc., enables mobile vehicles to predict an upcoming accident and alert the human drivers in time or proactively take precautionary actions to avoid the accident. The actualization of this concept requires the use of channel access protocols that can guarantee reliable and timely broadcast of safety messages. This paper investigates the application of network coding concept to increase content of every transmission and achieve improved broadcast reliability with less number of retransmission. In particular, we proposed Code Aided Retransmission-based Error Recovery (CARER) scheme, introduced an RTB/CTB handshake to overcome hidden node problem and reduce packets collision rate. In order to avoid broadcast storm problem associated with the use of RTB/CTB packet in a broadcast transmission, we developed a rebroadcasting metric used to successfully select a vehicle to rebroadcast the encoded message. The performance of CARER protocol is clearly shown with detailed theoretical analysis and further validated with simulation experiments.
    • Token passing techniques to support real time communications on WTPN used for industrial control applications

      Karimireddy, Thanmayee; Zhang, Sijing; Reddy, Manideep; University of Bedfordshire; VIT University (Institute of Electrical and Electronics Engineers Inc., 2017-10-26)
      Wireless Token Passing Networks (WTPN) have gained significant attention in the recent times due to the ability of these networks to overcome the challenge of channel access sharing faced by the wireless networks supporting real-time communications. A WTPN is specified as the network implementing token passing techniques in order to share the wireless channel access medium. In the token passing techniques, only the wireless node holding the token acquires the chance to access the channel, which thereby helps in offering guaranteed channel access time to the nodes connected in the WTPN. The efficiency of token passing techniques in offering guaranteed channel access when compared with master-slave and polling mechanisms resulted in the development of several token passing techniques. This paper intends in surveying and simulating the state-of-art token passing techniques that are applicable for supporting real-time communications over WTPN used for industrial control applications. The theoretical and practical comparison of the token passing techniques presented in this paper can help the practitioners in choosing the suitable token passing technique.
    • Topic-aware visual citation tracing via enhanced term weighting for efficient literature retrieval

      Zhao, Youbing; Wei, Hui; Wu, Shaopeng; Parvinzamir, Farzad; Deng, Zhikun; Zhao, Xia; Ersotelos, Nikolaos; Dong, Feng; Clapworthy, Gordon J.; Liu, Enjie; et al. (Springer Verlag, 2017-12-31)
      Efficient retrieval of scientific literature related to a certain topic plays a key role in research work. While little has been done on topic-enabled citation filtering in traditional citation tracing, this paper presents visual citation tracing of scientific papers with document topics taken into consideration. Improved term selection and weighting are employed for mining the most relevant citations. A variation of the TF-IDF scheme, which uses external domain resources as references is proposed to calculate the term weighting in a particular domain. Moreover document weight is also incorporated in the calculation of term weight from a group of citations. A simple hierarchical word weighting method is also presented to handle keyword phrases. A visual interface is designed and implemented to interactively present the citation tracks in chord diagram and Sankey diagram.
    • Towards 5G: a reinforcement learning-based scheduling solution for data traffic management

      Comşa, Ioan-Sorin; Zhang, Sijing; Aydin, Mehmet Emin; Kuonen, Pierre; Lu, Yao; Trestian, Ramona; Ghinea, Gheorghiţă; Brunel University; University of Bedfordshire; University of the West of England; et al. (IEEE, 2018-08-06)
      Dominated by delay-sensitive and massive data applications, radio resource management in 5G access networks is expected to satisfy very stringent delay and packet loss requirements. In this context, the packet scheduler plays a central role by allocating user data packets in the frequency domain at each predefined time interval. Standard scheduling rules are known limited in satisfying higher Quality of Service (QoS) demands when facing unpredictable network conditions and dynamic traffic circumstances. This paper proposes an innovative scheduling framework able to select different scheduling rules according to instantaneous scheduler states in order to minimize the packet delays and packet drop rates for strict QoS requirements applications. To deal with real-time scheduling, the Reinforcement Learning (RL) principles are used to map the scheduling rules to each state and to learn when to apply each. Additionally, neural networks are used as function approximation to cope with the RL complexity and very large representations of the scheduler state space. Simulation results demonstrate that the proposed framework outperforms the conventional scheduling strategies in terms of delay and packet drop rate requirements.
    • Towards DR inventor: a tool for promoting scientific creativity

      O’Donoghue, D.P.; Saggion, H.; Dong, Feng; Hurley, D.; Abgaz, Y.; Zheng, X.; Corcho, O,; Zhang, J.J.; Careil, J.M.; Mahdian, B.; et al. (Jozef Stefan Institute, 2014-12-31)
      We propose an analogy-based model to promote creative scientific reasoning among its users. Dr Inventor aims to find novel and potentially useful creative analogies between academic documents, presenting them to users as potential research questions to be explored and investigated. These novel comparisons will thereby drive its users’ creative reasoning. Dr Inventor is aimed at promoting Big-C Creativity and the H-creativity associated with true scientific creativity.
    • Towards sparse characterisation of on-body ultra-wideband wireless channels

      Yang, Xiaodong; Ren, Aifeng; Zhang, Zhiya; Ur-Rehman, Masood; Abbasi, Qammer Hussain; Alomainy, Akram; Xidian University; University of Bedfordshire; Texas A&M University at Qatar (IET, 2015-07-01)
      With the aim of reducing cost and power consumption of the receiving terminal, compressive sensing (CS) framework is applied to on-body ultra-wideband (UWB) channel estimation. It is demonstrated in this Letter that the sparse on-body UWB channel impulse response recovered by the CS framework fits the original sparse channel well; thus, on-body channel estimation can be achieved using low-speed sampling devices.
    • Tracking human motion direction with commodity wireless networks

      Rahaman, Habibur; Dyo, Vladimir; University of Bedfordshire (IEEE, 2021-09-07)
      Detecting when a person leaves a room, or a house is essential to create a safe living environment for people suffering from dementia or other mental disorders. The approaches based on wearable devices, e.g. GPS bracelets may detect such events require periodic maintenance to recharge or replace batteries, and therefore may not be suitable for certain types of users. On the other hand, camera-based systems require illumination and raise potential privacy concerns. In this paper, we propose a device-free walking direction detection approach based on RF-sensing, which does not require a person to wear any equipment. The proposed approach monitors the signal strength fluctuations caused by the human body on ambient wireless links and analyses its spatial patterns using a convolutional neural network to identify the walking direction. The approach has been evaluated experimentally to achieve up to 98% classification accuracy depending on the environment.
    • Tracking objects robot for healthcare environments

      Kanjaruek, Saranya; Li, Dayou; Khon Kaen University, Thailand; University of Bedfordshire (Institute of Electrical and Electronics Engineers Inc., 2018-02-01)
      There are many elder who are affected from Alzheimer's disease. Memory problems, confusion and forgetting objects, recent conversations or places are example of Alzheimer symptoms. Nursing home needs medical devices or robot to provide service for patients. Tracking Objects Robot traces and locates objects for the Alzheimer's patients to find objects in environment. This paper uses frequency and recency technique in order to locate the location of objects in dynamic environment by associating semantic knowledge of object with instance in ontology.
    • Transferring porous layer from InP wafer based on the disturbance

      Zhang, Yang; Cao, Liang; Chai, Xiangyu; Liang, Kaihua; Han, Yong-Lu; Wang, Yanqi; Wang, Zuobin; Wang, Shuting; Weng, Zhankun; Changchun University of Science and Technology; et al. (Institute of Electrical and Electronics Engineers Inc., 2017-01-16)
      We present a new method to transfer the three dimensional (3D) porous layer from InP wafer based on the disturbance, during electrochemical etching of n-InP (100) with chronopotentiometry with current ramp in 3 mol-L-1 NaCl solution. The potential bursting phenomenon was observed due to the disturbing instantaneously. In addition, the correlation between the amplitude of the potential and the porous layers separated from the InP wafer was discussed.
    • Tri-band millimetre-wave antenna for body-centric networks

      Ur-Rehman, Masood; Adekanye, Michael; Chattha, Hassan Tariq; University of Bedfordshire; Islamic University Madinah (Elsevier, 2018-04-03)
      This paper presents design of a tri-band slotted patch antenna operating at millimetre-wave frequencies of 28 GHz, 38 GHz and 61 GHz. The proposed antenna carries an overall size of 5.1mm×5mm×0.254mm employing a single layer, slotted patch structure combining L- and F-shaped slots. It is excited by a single-feed microstrip line. The antenna is tested in free space as well as in wearable configurations and results show that it offers a good impedance matching, sufficient -10 dB bandwidth and wide radiation coverage at the three bands of interest effectively countering the effects of human body presence. It achieves a peak gain of 7.2 dBi in off-body and 8.3 dBi in on-body configuration. Minimum efficiency values are observed to be 85% in off-body while 54% in on-body scenarios. A comparative analysis with published relevant work shows that the proposed antenna is inexpensive, easy to integrate and works efficiently in tri-band wearable and implantable arrangements. These features make it a good candidate for current and future applications of Body-centric Networks operating at millimetre-wave ranges.
    • Trustworthiness in the patient centred health care system

      Liu, Enjie; Feng, Xiaohua; University of Bedfordshire (Springer Verlag, 2014-06-27)
      The trend of the future health care system is patient centred, and patients' involvement is a key to success. ICT will play an important role in enabling and helping patients or citizens to manage and communicate on the individual's health related issues. This includes private and confidential information. Trustworthiness is therefore one of the most vital aspects in such systems. This paper first presents the prototype structure of the health care system, and then discusses questions regarding the trustworthiness of the system. © Springer-Verlag Berlin Heidelberg 2014.
    • Tunable electrochemical oscillation and regular 3D nanopore arrays of InP

      Chai, Xiangyu; Weng, Zhankun; Xu, Liping; Wang, Zuobin; Changchun University of Science and Technology; University of Bedfordshire (Electrochemical Society Inc., 2015-06-16)
      Tunable potential oscillations are obtained by electrochemical etching of n-InP (100) in the 3MNaCl solution using the chronoptentiometry with current ramp. The regular 3D nanopore arrays are formed with the change of the current density from 320 to 260 mA · cm-2 at the scan rate 0.72 - 0.80 mA · cm-2 · s-1. The results showed that the current density ranges and scan rate have the effect on the E-t curves and the pore's morphology. The scan rate can regulate not only on the charge consumed per period but also on the amplitude of potential oscillation, and shown that the charge per period and the amplitude can be tuned when proper electrochemical parameters are selected. Furthermore, the pore's morphology will change from the regular structure to irregular with the increasing of the scan rate. In addition, the relation has also been discussed between the E-t curves and the pore's morphology.