Table of Contents
Computer systems are usually run using binary logic for their operations but according to Acampora and Loia (2011), fuzzy logic can be used to describe a complex system. Zadeh (1996) explains that fuzzy logic is computing in which approximations are used rather than absolute or definite values which is unlike Boolean logic where definite values are used. He continues to say that fuzzy logic uses a mix of natural languages, or human speech, and computing with fuzzy variables. In order to make proper use of fuzzy logic a new language has been made to ease and simplify the modelling of fuzzy systems which is the Fuzzy Markup Language based on eXtensible Markup Language (XML) (Acampora and Loia, 2011). This report will describe the Fuzzy Markup Language, provide examples of its applications, and the prospects of this field.
According to Almaraashi (2017), fuzzy logic systems, which are based on the theory of fuzzy logic proposed by Zadeh in 1965, have been used to represent information in a familiar human way of thinking. As an XML based language, Fuzzy Markup Language describes a fuzzy logic system as well as facilitates the representation of a fuzzy control system (Acampora and Loia, 2011; Lee et al., 2018). Lee et al. (2018) continue to state that FML can produce cheap, fast, and robust solutions for complex problems using algorithms inspired by nature and population growth. Fuzzy logic has been applied in numerous fields. Singh et al (2013) say that fuzzy logic has been used in many fields such as, risk assessment, data anlysis, and optimisation.
Fuzzy Markup Language itself uses a set of syntaxes, tags, and attributes, to describe a fuzzy logic system (Acampora and Loia, 2011). Acampora and Loia (2011) say that FML uses the following main tags:
A dozen fuzzy shape tags
- A dozen fuzzy shape tags
The tag <FUZZYCONTROLLER> is the root tag of FML which means it is necessary to type in at the start of every FML program (Acampora and Loia, 2011). The main set of fuzzy concepts are contained in the <KNOWLEDGEBASE> tag. They continue to state that the previous tag uses a pair of tags which are <FUZZYVARIABLE> and <FUZZYTERM>. The fuzzy concept is determined by the <FUZZYVARIABLE> tag, and the <FUZZYTERM> tag, which is nested in the <FUZZYVARIABLE> tag, sets a term describing the fuzzy concept. Within the <FUZZYTERM> tag are a dozen fuzzy shape tags, these are used to determine how the model will be displayed, and the <POINT> tag is used to place the coordinates for the displayed shape (Acampora and Loia, 2011). All the data entered in the FML can be displayed and visualised using a tool call VisualFMLTool.
One example of fuzzy logic systems application is the detection of potential fires in electric vehicles. Dattathreya, Singh and Meitzler (2012), proposed a fuzzy system (FDNCT) which monitors the engine for possible fire-starting incidents such as high-temperature, moisture and irregular voltages. If such an incident is found, the system then proposes a possible course of action like spray fire-extinguishing agent, have the sprayer in standby, and alert the passengers of the vehicle. They continue to say that a fuzzy system is used since standard analytical methods cannot handle specific linguistic terms. The FDNCT demonstrated more favourable results over a similar approach called the singleton (Dattathreya, Singh and Meitzler 2012). To perform an action for the engine compartment, the FDNCT had an average of 0.09 seconds while the singleton approach had an average of 0.12 seconds. The time it takes to act has been reduced by 25%. The battery compartment had similar results in which the FDNCT had an average of 0.11 seconds to execute a command while the singleton had an average of 0.19 seconds to do the same action. This reduces the action time by approximately 42% for the FDNCT.
A final example of FML is Ontology Web Language for Fuzzy Control (OWL-FC), which is a Fuzzy Markup Language based on OWL. This language was introduced to model the fuzziness of the real world by De Maio et al. (2012). OWL-FC has three main classes associated with it they are FuzzyControl Profile, and Model, each of these classes have their own properties. Each Fuzzy Control declaration has its own reference class that is represented by the FuzzyControl class. De Maio et al. (2012) expressed that what has been done by the FuzzyControl is described by the Profile class using the properties Presents and PresentedBy. The Presents property means that the FuzzyControl is represented by the Profile while PresentedBy is the inverse of that. The class Model on the other hand describes how the FuzzyControl class works. Similarly, it has two properties it can use. The properties are Describes and DescribedBy. Described is a Model’s description of a FuzzyControl and DescribedBy is the inverse of that (De Maio et al., 2012). One possible application of OWL-FC is landslide risk prevention and detection. In this example, De Maio et al. (2012) modelled the possibility of a landslide as the intersection of the fuzzy concepts, degree of rainfall and humidity. They continue to say that one of the characteristics that differentiate OWL-FC from FML is its automated usage of fuzzy controllers. Since it is based on OWL, the data entered is meant to be processed rather than directly read. Although OWL-FC can be used in such areas, it falls short in others especially when the information given is imprecise.
FML could potentially be used in education. Lee et al. (2018) experimented with the application of Particle Swarm Optimisation (PSO)-based Fuzzy Markup Language. They used this method (PFML) to evaluate student learning performance. To make the results of the experiment close in to the required results, PFML was used to optimise the knowledge base of FML. This resulted in favourable results for the experiment. Student results were estimated in four levels, and this can be used to propose improvement methods for them. Lee et al. (2018) acknowledge some weakness with this method. One of which is to propose learning recommendations for students of different levels.
Another prospect for FML is predicting occurrences of solar energy. Almaraashi (2017) conducted an experiment using fuzzy logic systems to find the availability of solar energy in 8 locations in Saudi Arabia for the next day.The system built the model using 10 meteorological variables, The first experiment conducted had an accuracy rating of 79.75% while the second modified model had an accuracy of 88% (Almaraashi, 2017). In order to increase the accuracy of the model he proposes the addition of more related variables to enhance the ratings further.
Fuzzy Markup Language, an XML based language, introduces a new method to model fuzzy logic and fuzzy systems. Fuzzy logic is a method of computing where computers can process vague information, as Zadeh (1996) calls it computing with words. FML allows the modelling of fuzzy logic systems to be done easily and productively (Acampora and Lee, 2012; Acampora and Loia, 2011). As Acampora and Loia (2011) state, FML has its own set of syntaxes, tags, and attributes to describe a fuzzy system. FML assists in describing a fuzzy logic system so it can be used anywhere a fuzzy system is used such as risk assessment, data analysis, and optimisation. According to Lee et al. (2018) one possible prospect for FML is its use in education performance evaluation. This could allow education systems to administer proper resources towards the need of their students. Another achievable prospect is its use to predict the availability of solar energy (Almaraashi, 2017). Almaraashi (2017) used a fuzzy system to predict the availability of solar energy the next day, and his experiments produced favourable results regarding its accuracy which can be further enhanced.
- Acampora, G. and Loia, V., 2011. Fuzzy Markup Language: A New Solution for Transparent Intelligent Agents. IEEE Symposium on Intelligent Agent (IA) [online], 1-6. Available at: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5953621&isnumber=5953603 [Accessed 14 November 2018]
- Acampora, G. and Lee, C., 2012. Special issue on fuzzy ontologies and fuzzy markup language applications. Soft Computing [online], 16(7) (July), 1107-1108. Available at: https://link.springer.com/article/10.1007/s00500-011-0785-1 [Accessed 21 November 2018]
- Almaraashi, M., (2017). Short-term prediction of solar energy in Saudi Arabia using automated-design fuzzy logic systems. PLoS ONE [online], 12(8) (August), 1-16. Available at: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0182429 [Accessed 14 November 2018]
- Dattathreya, M.S., Singh, H., and Meitzler, T., 2012. Detection and Elimination of a Potential Fire in Engine and Battery Compartments of Hybrid Electric Vehicles. Advances in Fuzzy Systems [online], 12, Article ID 687652, 1-11, Available at: https://www.hindawi.com/journals/afs/2012/687652/ [Accessed 14 November 2018]
- De Maio, C., Fenza, G., Furno, D., Loia, V., and Senatore, S., 2012. OWL-FC: An upper ontology for semantic modelling of Fuzzy Control. Soft Computing [online], 16(7) (July), 1153-1164. DOI: 10.1007/s00500-011-0790-4 [21 November 2018]
- Lee, C., Wang, M., Wang, C., Teytaud, O., Liu, J., Lin, S., and Hung, P., 2018. PSO-Based Fuzzy Markup Language for student Learning Performance Evaluation and Educational Application. IEEE Transactions on Fuzzy Systems [online], 26(5) (October), 2618-2633. Available at: https://ieeexplore.ieee.org/document/8304801 [Accessed 12 November 2018]
- Zadeh, L.A., 1996. Fuzzy Logic = Computing with Words. IEEE Transactions on Fuzzy Systems [online], 4(2) (May), 103-111. Available at: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=493904&isnumber=10655 [Accessed 12 November 2018]
Cite This Work
To export a reference to this article please select a referencing stye below:
Related ServicesView all
DMCA / Removal Request
If you are the original writer of this essay and no longer wish to have your work published on UKEssays.com then please: