Also maintained at Google Scholar

Journals

  1. Matthews, S.G., Miller, A.L., Clapp, J., Plötz, T. & Kyriazakis, I. (2016) Early detection of health and welfare compromises through automated detection of behavioural changes in pigs. The Veterinary Journal. 21743–51.

    Abstract Early detection of health and welfare compromises in commercial piggeries is essential for timely intervention to enhance treatment success, reduce impact on welfare, and promote sustainable pig production. Behavioural changes that precede or accompany subclinical and clinical signs may have diagnostic value. Often referred to as sickness behaviour, this encompasses changes in feeding, drinking, and elimination behaviours, social behaviours, and locomotion and posture. Such subtle changes in behaviour are not easy to quantify and require lengthy observation input by staff, which is impractical on a commercial scale. Automated early-warning systems may provide an alternative by objectively measuring behaviour with sensors to automatically monitor and detect behavioural changes. This paper aims to: (1) review the quantifiable changes in behaviours with potential diagnostic value; (2) subsequently identify available sensors for measuring behaviours; and (3) describe the progress towards automating monitoring and detection, which may allow such behavioural changes to be captured, measured, and interpreted and thus lead to automation in commercial, housed piggeries. Multiple sensor modalities are available for automatic measurement and monitoring of behaviour, which require humans to actively identify behavioural changes. This has been demonstrated for the detection of small deviations in diurnal drinking, deviations in feeding behaviour, monitoring coughs and vocalisation, and monitoring thermal comfort, but not social behaviour. However, current progress is in the early stages of developing fully automated detection systems that do not require humans to identify behavioural changes; e.g., through automated alerts sent to mobile phones. Challenges for achieving automation are multifaceted and trade-offs are considered between health, welfare, and costs, between analysis of individuals and groups, and between generic and compromise-specific behaviours.

    @article{matthews-2016-tvj,
      title = {Early detection of health and welfare compromises through automated detection of behavioural changes in pigs},
      journal = {The Veterinary Journal},
      volume = {217},
      number = {},
      pages = {43--51},
      year = {2016},
      issn = {1090-0233},
      doi = {10.1016/j.tvjl.2016.09.005},
      url = {http://www.sciencedirect.com/science/article/pii/S1090023316301538},
      author = {Matthews, Stephen G. and Miller, Amy L. and Clapp, James and Pl{\"o}tz, Thomas and Kyriazakis, Ilias}
    }
    
  2. Matthews, S.G., Gongora, M.A. & Hopgood, A.A. (2013) Evolutionary Algorithms and Fuzzy Sets for Discovering Temporal Rules. Applied Mathematics and Computer Science. 23 (4), 855–868.

    A novel method is presented for mining fuzzy association rules that have a temporal pattern. Our proposed method contributes towards discovering temporal patterns that could otherwise be lost from defining the membership functions before the mining process. The novelty of this research lies in exploring the composition of fuzzy and temporal association rules, and using a multi-objective evolutionary algorithm combined with iterative rule learning to mine many rules. Temporal patterns are augmented into a dataset to analyse the method’s ability in a controlled experiment. It is shown that the method is capable of discovering temporal patterns, and the effect of Boolean itemset support on the efficacy of discovering temporal fuzzy association rules is presented.

    @article{matthews-2013-amcs,
      author = {Matthews, Stephen G. and Gongora, Mario A. and Hopgood, Adrian A.},
      title = {Evolutionary Algorithms and Fuzzy Sets for Discovering Temporal Rules},
      journal = {Applied Mathematics and Computer Science},
      year = {2013},
      volume = {23},
      pages = {855--868},
      number = {4},
      doi = {10.2478/amcs-2013-0064}
    }
    
  3. Matthews, S.G., Gongora, M.A. & Hopgood, A.A. (2013) Web Usage Mining with Evolutionary Extraction of Temporal Fuzzy Association Rules. Knowledge-Based Systems. 54 (0), 66–72.

    In Web usage mining, fuzzy association rules that have a temporal property can provide useful knowledge about when associations occur. However, there is a problem with traditional temporal fuzzy association rule mining algorithms. Some rules occur at the intersection of fuzzy sets’ boundaries where there is less support (lower membership), so the rules are lost. A genetic algorithm (GA)-based solution is described that uses the flexible nature of the 2-tuple linguistic representation to discover rules that occur at the intersection of fuzzy set boundaries. The GA-based approach is enhanced from previous work by including a graph representation and an improved fitness function. A comparison of the GA-based approach with a traditional approach on real-world Web log data discovered rules that were lost with the traditional approach. The GA-based approach is recommended as complementary to existing algorithms, because it discovers extra rules.

    @article{matthews-2013-kbs,
      author = {Matthews, Stephen G. and Gongora, Mario A. and Hopgood, Adrian A.},
      title = {Web Usage Mining with Evolutionary Extraction of Temporal Fuzzy Association Rules},
      journal = {Knowledge-Based Systems},
      year = {2013},
      volume = {54},
      pages = {66--72},
      number = {0},
      doi = {10.1016/j.knosys.2013.09.003},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Web-Usage-Mining-with-Evolutionary-Extraction-of-Temporal-Fuzzy-Association-Rules.pdf}
    }
    
  4. Elizondo, D.A. & Matthews, S.G. (2008) Recent Patents on Computational Intelligence. Recent Patents on Computer Science. 1 (2), 110–117.

    The field of computational intelligence is the successor of artificial intelligence. It combines elements of learning, evolution and fuzzy logic to create programs that can exhibit, to some degree, intelligent behaviour. The last few years have seen a growth in theoretical and practical developments in this field. This paper presents an extensive survey of the latest patents in the domain of computational intelligence. It discusses and summarises some of these latest developments in terms of patents. The patents are categorised by their domain of application: Artificial Neural Networks, Fuzzy Logic and Evolutionary Algorithms and further classified by the application area.

    @article{matthews-2008-rpcs,
      author = {Elizondo, David A. and Matthews, Stephen G.},
      title = {Recent Patents on Computational Intelligence},
      journal = {Recent Patents on Computer Science},
      year = {2008},
      volume = {1},
      pages = {110--117},
      number = {2},
      month = jun,
      doi = {10.2174/2213275910801020110},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Recent-Patents-on-Computational-Intelligence.pdf}
    }
    

Magazines

  1. Matthews, S.G. (2011) What it means to be a young CI researcher in the 21st century. IEEE Computational Intelligence Magazine. 6 (3), 6–7.

    (This article won the IEEE CIS GOLD essay competition and was accepted for publication. See http://www.cci.dmu.ac.uk/news-archive/186-stephen-matthews-wins-ieee-cis-essay-competition)

    @article{matthews-2011-cim,
      author = {Matthews, Stephen G.},
      title = {What it means to be a young {CI} researcher in the 21st century},
      journal = {{IEEE} Computational Intelligence Magazine},
      year = {2011},
      volume = {6},
      pages = {6--7},
      number = {3},
      doi = {10.1109/MCI.2011.941582},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/What-it-means-to-be-a-young-CI-researcher-in-the-21st-century.pdf},
      note = {Won IEEE CIS essay competition (http://ewh.ieee.org/r8/ukri/cis/)}
    }
    

Conferences

  1. Wang, Y., Matthews, S.G. & Bryson, J.J. (2014) 'Evolving Evolvability in the Context of Environmental Change: A Gene Regulatory Network (GRN) Approach', in Proceedings of ALIFE 14: The 14th International Conference on the Synthesis and Simulation of Living Systems. [Online]. July 2014 New York: The MIT Press. pp. 47–53.

    Evolvability is the capacity of a genotype to rapidly adjust to certain types of environmental challenges or opportunities. This capacity, documented in nature, reflects foresight enabled by the capacity of evolution to capture and represent regularities not only in extant environments, but in the ways in which the environments tend to change. Here we posit that evolvability substantially benefits from the hierarchical representations afforded by Gene Regulatory Networks (GRNs). We present an extension of standard Genetic Algorithms (GAs) and demonstrate its capacity to learn a genotype phylogeny able to express rapid phenotypic shifts in the context of an oscillating environment.

    @inproceedings{matthews-2014-alife,
      author = {Wang, Yifei and Matthews, Stephen G. and Bryson, Joanna J},
      title = {Evolving Evolvability in the Context of Environmental Change: A Gene Regulatory Network (GRN) Approach},
      booktitle = {Proceedings of ALIFE 14: The 14th International Conference on the Synthesis and Simulation of Living Systems},
      year = {2014},
      pages = {47--53},
      address = {New York},
      month = jul,
      publisher = {The MIT Press},
      doi = {10.7551/978-0-262-32621-6-ch010},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Evolving-Evolvability-in-the-Context-of-Environmental-Change_-A-Gene-Regulatory-Network-(GRN)-Approach.pdf}
    }
    
  2. Matthews, S.G. & Martin, T.P. (2014) 'Possibilistic Projected Categorical Clustering via Cluster Cores', in Proceedings of The 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2014). [Online]. July 2014 Beijing: IEEE. pp. 1–8.

    Projected clustering discovers clusters in subsets of locally relevant attributes. There is uncertainty and imprecision about how groups of categorical values are learnt from data for projected clustering and also the data itself. A method is presented for learning discrete possibility distributions of categorical values from data for projected clustering in order to model uncertainty and imprecision. Empirical results show that fewer, more accurate, more compact, and new clusters can be discovered by using possibility distributions of categorical values when compared to an existing method based on Boolean memberships. This potentially allows for new relationships to be identified from data.

    @inproceedings{matthews-2014-fuzz-a,
      author = {Matthews, Stephen G. and Martin, Trevor P.},
      title = {Possibilistic Projected Categorical Clustering via Cluster Cores},
      booktitle = {Proceedings of The 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2014)},
      year = {2014},
      pages = {1--8},
      address = {Beijing},
      month = jul,
      publisher = {IEEE},
      doi = {10.1109/FUZZ-IEEE.2014.6891764},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Possibilistic-Projected-Categorical-Clustering-via-Cluster-Cores.pdf}
    }
    
  3. Matthews, S.G. (2014) 'Tuning Larger Membership Grades for Fuzzy Association Rules', in Proceedings of The 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2014). [Online]. July 2014 Beijing: IEEE. pp. 1–8.

    Sigma count measures scalar cardinality of fuzzy sets. A problem with sigma count is that values of scalar cardinality are calculated entirely from many small membership grades or entirely from few large membership grades. Two novel scalar cardinality measures are proposed for the fitness of a genetic algorithm for tuning membership functions prior to fuzzy association rule mining so that individual membership grades are larger. Preliminary results show a decrease in small membership grades and an increase in large membership grades for fuzzy association rules tested on real-world benchmark datasets.

    @inproceedings{matthews-2014-fuzz-b,
      author = {Matthews, Stephen G.},
      title = {Tuning Larger Membership Grades for Fuzzy Association Rules},
      booktitle = {Proceedings of The 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2014)},
      year = {2014},
      pages = {1--8},
      address = {Beijing},
      month = jul,
      publisher = {IEEE},
      doi = {10.1109/FUZZ-IEEE.2014.6891765},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Tuning-Larger-Membership-Grades-for-Fuzzy-Association-Rules.pdf}
    }
    
  4. Coupland, S. & Matthews, S.G. (2013) 'Using nonstationary fuzzy sets to improve the tractability of fuzzy association rules', in Proceedings of The 2013 IEEE Symposium on Advances in Type-2 Fuzzy Logic Systems (T2FUZZ). [Online]. April 2013 Singapore: IEEE. pp. 9–14.

    Modern organisations now collect very large volumes of data about customers, suppliers and other factors which may impact upon their business. There is a clear need to be able to mine this data and present it to decision makers in a clear and coherent manner. Fuzzy association rules are a popular method to identifying important and meaningful relationships within large data sets. Recently a fuzzy association rule has been proposed that uses the 2-tuple linguistic representation. This paper presents a methodology which makes use of non-stationary fuzzy sets to post process 2-tuple fuzzy association rules reducing the size of the mined rule set by around 20% whilst retaining the semantic meaning of the rule set.

    @inproceedings{matthews-2013-t2fuzz,
      author = {Coupland, Simon and Matthews, Stephen G.},
      title = {Using nonstationary fuzzy sets to improve the tractability of fuzzy association rules},
      booktitle = {Proceedings of The 2013 IEEE Symposium on Advances in Type-2 Fuzzy Logic Systems (T2FUZZ)},
      year = {2013},
      pages = {9--14},
      address = {Singapore},
      month = apr,
      publisher = {IEEE},
      doi = {10.1109/T2FZZ.2013.6613293},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Using-Nonstationary-Fuzzy-Sets-to-Improve-the-Tractability-of-Fuzzy-Association-Rules.pdf}
    }
    
  5. Rowlingson, R., Healing, A., Shittu, R., Matthews, S.G. & Ghanea-Hercock, R. (2013) 'Visual Analytics in the Cyber Security Operations Centre', in Proceedings of The Information Systems Technology Panel Symposium on Visual Analytics (IST-116/RSY-028). [Online]. October 2013 Shrivenham: NATO.

    Skilled cyber security experts are becoming the last line of defence in roles such as analysts in a Security Operations Centre (SOC). The skills and knowledge of analysts are clearly critical, and in short supply, so the right resources to support their motivations for network monitoring and alerting, can enable more effective use of analysts’ time. Visual Analytics (VA) can support the motivation of such individuals in inter alia, detecting, investigating and assessing cyber threat. The requirements and application of VA in a cyber SOC are described in the context of previous research in this area. The Saturn tool, developed by and in use at BT, is described and an overview of its deployment and feedback from analysts is provided. Potential future developments for Saturn, and other VA tools for cyber SOCs, are also outlined.

    @inproceedings{matthews-2013-nato,
      author = {Rowlingson, Robert and Healing, Alex and Shittu, Riyanat and Matthews, Stephen G. and Ghanea-Hercock, Robert},
      title = {Visual Analytics in the Cyber Security Operations Centre},
      booktitle = {Proceedings of The Information Systems Technology Panel Symposium on Visual Analytics (IST-116/RSY-028)},
      year = {2013},
      pages = {},
      address = {Shrivenham},
      month = oct,
      publisher = {NATO},
      note = {Won best paper award}
    }
    
  6. Matthews, S.G., Gongora, M.A., Hopgood, A.A. & Ahmadi, S. (2012) 'Temporal Fuzzy Association Rule Mining with 2-tuple Linguistic Representation', in Proceedings of The 2012 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2012). [Online]. June 2012 Brisbane: IEEE. pp. 1–8.

    This paper reports on an approach that contributes towards the problem of discovering fuzzy association rules that exhibit a temporal pattern. The novel application of the 2-tuple linguistic representation identifies fuzzy association rules in a temporal context, whilst maintaining the interpretability of linguistic terms. Iterative Rule Learning (IRL) with a Genetic Algorithm (GA) simultaneously induces rules and tunes the membership functions. The discovered rules were compared with those from a traditional method of discovering fuzzy association rules and results demonstrate how the traditional method can loose information because rules occur at the intersection of membership function boundaries. New information can be mined from the proposed approach by improving upon rules discovered with the traditional method and by discovering new rules.

    @inproceedings{matthews-2012-fuzz,
      author = {Matthews, Stephen G. and Gongora, Mario A. and Hopgood, Adrian A. and Ahmadi, Samad},
      title = {Temporal Fuzzy Association Rule Mining with 2-tuple Linguistic Representation},
      booktitle = {Proceedings of The 2012 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2012)},
      year = {2012},
      pages = {1--8},
      address = {Brisbane},
      month = jun,
      publisher = {IEEE},
      doi = {10.1109/FUZZ-IEEE.2012.6251173},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Temporal-Fuzzy-Association-Rule-Mining-with-2-tuple-Linguistic-Representation.pdf}
    }
    
  7. Matthews, S.G., Gongora, M. & Hopgood, A. (2011) 'Evolving Temporal Fuzzy Itemsets from Quantitative Data with a Multi-Objective Evolutionary Algorithm', in The IEEE 5th International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS 2011). [Online]. April 2011 Paris: IEEE. pp. 9–16.

    We present a novel method for mining itemsets that are both quantitative and temporal, for association rule mining, using multi-objective evolutionary search and optimisation. This method successfully identifies temporal itemsets that occur more frequently in areas of a dataset with specific quantitative values represented with fuzzy sets. Current approaches preprocess data which can often lead to a loss of information. The novelty of this research lies in exploring the composition of quantitative and temporal fuzzy itemsets and the approach of using a multi-objective evolutionary algorithm. This preliminary work presents the problem, a novel approach and promising results that will lead to future work. Results show the ability of NSGA-II to evolve target itemsets that have been augmented into synthetic datasets. Itemsets with different levels of support have been augmented to demonstrate this approach with varying difficulties.

    @inproceedings{matthews-2011-gefs,
      author = {Matthews, Stephen G. and Gongora, Mario and Hopgood, Adrian},
      title = {Evolving Temporal Fuzzy Itemsets from Quantitative Data with a Multi-Objective Evolutionary Algorithm},
      booktitle = {The IEEE 5th International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS 2011)},
      year = {2011},
      pages = {9--16},
      address = {Paris},
      month = apr,
      publisher = {IEEE},
      doi = {10.1109/GEFS.2011.5949497},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Evolving-Temporal-Fuzzy-Itemsets-from-Quantitative-Data-with-a-Multi-Objective-Evolutionary-Algorithm.pdf}
    }
    
  8. Matthews, S.G., Gongora, M. & Hopgood, A. (2011) 'Evolving Temporal Fuzzy Association Rules from Quantitative Data with a Multi-Objective Evolutionary Algorithm', in Emilio Corchado et al. (eds.) Hybrid Artificial Intelligent Systems (Proceedings of HAIS 2011). Lecture Notes in Computer Science. [Online]. Springer Berlin / Heidelberg. pp. 198–205.

    A novel method for mining association rules that are both quantitative and temporal using a multi-objective evolutionary algorithm is presented. This method successfully identifies numerous temporal association rules that occur more frequently in areas of a dataset with specific quantitative values represented with fuzzy sets. The novelty of this research lies in exploring the composition of quantitative and temporal fuzzy association rules and the approach of using a hybridisation of a multi-objective evolutionary algorithm with fuzzy sets. Results show the ability of a multi-objective evolutionary algorithm (NSGA-II) to evolve multiple target itemsets that have been augmented into synthetic datasets.

    @incollection{matthews-2011-hais,
      author = {Matthews, Stephen G. and Gongora, Mario and Hopgood, Adrian},
      title = {Evolving Temporal Fuzzy Association Rules from Quantitative Data with a Multi-Objective Evolutionary Algorithm},
      booktitle = {Hybrid Artificial Intelligent Systems (Proceedings of HAIS 2011)},
      publisher = {Springer Berlin / Heidelberg},
      year = {2011},
      editor = {Corchado, Emilio and Kurzynski, Marek and Wozniak, Michal},
      volume = {6678},
      series = {Lecture Notes in Computer Science},
      pages = {198--205},
      month = may,
      doi = {10.1007/978-3-642-21219-2_26},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Evolving-Temporal-Fuzzy-Association-Rules-from-Quantitative-Data-with-a-Multi-Objective-Evolutionary-Algorithm.pdf}
    }
    
  9. Carter, J., Matthews, S. & Coupland, S. (2011) 'Teaching Robotics at the Postgraduate Level: Assessment and Feedback for On Site and Distance Learning', in Proceedings of The 2nd International Robotics in Education Conference (RIE 2011). [Online]. September 2011 Vienna: INNOC - Austrian Society for Innovative Computer Sciences. pp. 171–176.

    The MSc Intelligent Systems (IS) and the MSc Intelligent Systems and Robotics (ISR) programmes at De Montfort University are Masters level courses that are delivered both on-site and by distance learning. The courses have been running successfully on-site for 7 years and are now in the fourth year with a distance learning mode. Delivering material at a distance, especially where there is technical and practical content, always presents a challenge but the need to deliver a robotics module increased the challenges we faced significantly. There are two robotics modules though the second one is only available to those on MSc ISR. We have chosen to make the first robotics module, Mobile Robots, the focus of this paper because it was the first that had to be delivered and it is delivered to students on both programmes. This paper describes the assessment of students’ work and the subsequent feedback given to students within the course as a whole and more specifically, the Mobile Robots module. The approaches maximise the use of electronic methods and as such there is a specific focus on those students that are studying in distance learning mode. We believe it serves as a model for others attempting to assess students studying robotics courses at a distance.

    @inproceedings{matthews-2011-rie,
      author = {Carter, Jenny and Matthews, Stephen and Coupland, Simon},
      title = {Teaching Robotics at the Postgraduate Level: Assessment and Feedback for On Site and Distance Learning},
      booktitle = {Proceedings of The 2nd International Robotics in Education Conference
        (RIE 2011)},
      year = {2011},
      pages = {171--176},
      address = {Vienna},
      month = sep,
      publisher = {INNOC - Austrian Society for Innovative Computer Sciences},
      url = {http://www.innoc.at/fileadmin/user_upload/_temp_/RiE/Proceedings/23.pdf}
    }
    
  10. Matthews, S.G., Gongora, M.A. & Hopgood, A.A. (2010) 'Evolving Temporal Association Rules with Genetic Algorithms', in Max Bramer et al. (eds.) Research and Development in Intelligent Systems XXVII (Proceedings of AI-2010). [Online]. Cambridge: Springer London. pp. 107–120.

    A novel framework for mining temporal association rules by discovering itemsets with a genetic algorithm is introduced. Metaheuristics have been applied to association rule mining, we show the efficacy of extending this to another variant - temporal association rule mining. Our framework is an enhancement to existing temporal association rule mining methods as it employs a genetic algorithm to simultaneously search the rule space and temporal space. A methodology for validating the ability of the proposed framework isolates target temporal itemsets in synthetic datasets. The Iterative Rule Learning method successfully discovers these targets in datasets with varying levels of difficulty.

    @incollection{matthews-2010-ai,
      author = {Matthews, Stephen G. and Gongora, Mario A. and Hopgood, Adrian A.},
      title = {Evolving Temporal Association Rules with Genetic Algorithms},
      booktitle = {Research and Development in Intelligent Systems XXVII (Proceedings
      	of AI-2010)},
      publisher = {Springer London},
      year = {2010},
      editor = {Bramer, Max and Petridis, Miltos and Hopgood, Adrian},
      pages = {107--120},
      address = {Cambridge},
      month = dec,
      doi = {10.1007/978-0-85729-130-1_8},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/Evolving-Temporal-Association-Rules-with-Genetic-Algorithms.pdf}
    }
    
  11. Matthews, S.G., Coupland, S. & Zhou, S.-M. (2008) 'An Integrated Stereo Vision and Fuzzy Logic Controller for Following Vehicles in an Unstructured Environment', in The 2008 UK Workshop on Computational Intelligence (UKCI 2008). [Online]. September 2008 Leicester: De Montfort University. pp. 135–140.

    This paper demonstrates the concept of adaptive cruise control and vehicle following where by a safe distance is maintained between vehicles. A follower robot identifies a leader by using a stereo vision camera that determines the distance between the robots. The speed and direction of the robot are controlled by fuzzy logic controllers. The system performs considerably well in unstructured environments and exhibits good straight line performance. We put forward our framework for comparative analysis with alternative controllers.

    @inproceedings{matthews-2008-ukci,
      author = {Matthews, Stephen G. and Coupland, Simon and Zhou, Shang-Ming},
      title = {An Integrated Stereo Vision and Fuzzy Logic Controller for Following Vehicles in an Unstructured Environment},
      booktitle = {The 2008 UK Workshop on Computational Intelligence (UKCI 2008)},
      year = {2008},
      pages = {135--140},
      address = {Leicester},
      month = sep,
      publisher = {De Montfort University},
      url = {https://www.staff.ncl.ac.uk/stephen.matthews/pdf/An-Integrated-Stereo-Vision-and-Fuzzy-Logic-Controller-for-Following-Vehicles-in-an-Unstructured-Environment.pdf}
    }
    

PhD Thesis

  1. Matthews, S.G. (2012) Learning Lost Temporal Fuzzy Association Rules. PhD thesis thesis. [Online]. De Montfort University.

    Fuzzy association rule mining discovers patterns in transactions, such as shopping baskets in a supermarket, or Web page accesses by a visitor to a Web site. Temporal patterns can be present in fuzzy association rules because the underlying process generating the data can be dynamic. However, existing solutions may not discover all interesting patterns because of a previously unrecognised problem that is revealed in this thesis. The contextual meaning of fuzzy association rules changes because of the dynamic feature of data. The static fuzzy representation and traditional search method are inadequate. The Genetic Iterative Temporal Fuzzy Association Rule Mining (GITFARM) framework solves the problem by utilising flexible fuzzy representations from a fuzzy rule-based system (FRBS). The combination of temporal, fuzzy and itemset space was simultaneously searched with a genetic algorithm (GA) to overcome the problem. The framework transforms the dataset to a graph for efficiently searching the dataset. A choice of model in fuzzy representation provides a trade-off in usage between an approximate and descriptive model. A method for verifying the solution to the hypothesised problem was presented. The proposed GA-based solution was compared with a traditional approach that uses an exhaustive search method. It was shown how the GA-based solution discovered rules that the traditional approach did not. This shows that simultaneously searching for rules and membership functions with a GA is a suitable solution for mining temporal fuzzy association rules. So, in practice, more knowledge can be discovered for making well-informed decisions that would otherwise be lost with a traditional approach.

    @phdthesis{matthews-2012-phd,
      title = {Learning Lost Temporal Fuzzy Association Rules},
      author = {Matthews, Stephen G.},
      year = {2012},
      school = {De Montfort University},
      url = {https://www.dora.dmu.ac.uk/bitstream/handle/2086/8257/thesis.pdf}
    }