A Novel Genetic Fuzzy Markup Language and Its Application to Healthy Diet Assessment

Abstract

In this paper, we present a novel Genetic Fuzzy Markup Language (GFML)-based genetic fuzzy system, including the genetic learning base, the knowledge base and rule base of FML, the fuzzy inference engine, and the genetic learning mechanism. The GFML is applied to the genetic fuzzy system for dealing with the knowledge base, the rule base, and the genetic learning base of the healthy diet domain, including the ingredients and the contained servings of six food categories of some common food in Taiwan. Moreover, the proposed novel system is able to infer the healthy status of human's daily eating. In the proposed system, the domain experts first define the nutrient facts of the common food to construct the fuzzy food ontology. Meanwhile, the involved Taiwanese students of National University of Tainan (NUTN) record their daily meals for a constant period of time. Then, based on the built fuzzy profile ontology, fuzzy food ontology, and fuzzy personal food ontology, a GFML-based genetic fuzzy system is carried out to infer the possibility of dietary healthy level for one-day meals. The experimental results show that the proposed GFML-based genetic fuzzy system gives good results for the healthy diet assessment.

FML-based prediction agent and its application to game of Go

“…FML is with understandability, extendibility, and compatibility of implemented programs as well as efficiency of programming [10]. There are considerable research applications based on FML and Genetic FML (GFML) such as game of Go [12,13] and diet [14,15]. Additionally, Acampora et al [16] proposed FML Script to make it with evolving capabilities through a scripting language approach.…”

FML-based Japanese diet assessment system

“…As ontology development has been increasingly widespread and collaborative, developers have been creating ontologies by using different tools and languages. Therefore, the previous studies have employed ontology to incorporate with FML and fuzzy inference system [9,10,11,12]. For example, Lee et al proposed a fuzzy ontology to apply to diabetes decision support [13] and a type-2 fuzzy ontology to apply to the personal diabetic-diet recommendation [14] and diet assessment [9].…”

“…For example, Lee et al proposed a fuzzy ontology to apply to diabetes decision support [13] and a type-2 fuzzy ontology to apply to the personal diabetic-diet recommendation [14] and diet assessment [9]. Lee et al [10] also combined FML with the genetic algorithm to apply to healthy level assessment. Wang et al [11] proposed an ontology-based multi-agents and applied it to the intelligent healthcare.…”

“…What you burn through physical activity is energy out [15]. In order to help common people to easily meet the requirement in diet balance, such as choosing a wide variety of foods and drinks from all of the food groups as well as eating certain things in moderation [16], the previous works about assessing the dietary healthy level assessment for Taiwanese food [9,10,11] and Japanese food according to JFGST standard [12] have been published. JFGST standard uses "serving" to describe a certain nutrient value.…”

“…Next, based on the proposed method in previous work [10], an FML-based Japanese diet assessment system is also proposed to evaluate dietary healthy level of Japanese food intake for each day. Finally, according to the constructed Japanese food ontology, the proposed fuzzy inference mechanism infers the estimated dietary healthy level of the food intake.…”