Wastewater directly impacts individuals' health; hence, this topic is always a flame to motivate scientists to research treatment solutions. In terms of economy and infrastructure, building wastewater treatment systems is a significant and prominent challenge and problem in developing countries. Therefore, to construct a foundation for determining wastewater treatment solutions and building their management systems, we have proposed a lightweight ontology for wastewater treatment (WT) management. In particular, we concentrate on the keys of Biological Methodologies because these approaches are suitable for the environment and topography of Vietnam, especially at low cost. Accordingly, we pay attention to implementing this modeling by analyzing and classifying the primary biological and geographical features to serve in searching WT companies. Moreover, we leverage description logic that underlies the web ontology language OWL2-EL for representation, reasoning, and querying WT ontology-based modeling.
ganisms will continually metabolize organic matter by synthesizing new cell walls solely. During biological wastewater treatment, toxic substances in wastewater will be converted by microorganisms into harmless substances. pends on the individuals and that area’s living habits. vide a brief description of the popular biological methodTherefore, the wastewater treatment must suit each re- ologies. Moreover, we also put the structure of biologgion’s topography and mud (soil). ical wastewater treatment in this section. In Section 4,
In general, managing wastewater treatment in Viet- we discuss and illustrate how to implement an ontology nam is still challenging because individuals’ awareness through description logic EL for representation, reasonof waste disposal and the cost of operating and deploying ing, and query-answering. Finally, section 5 concludes treatment plants still need improvement. Moreover, the the paper. data management of wastewater treatment companies needs to be sucfiient and centralized. Therefore, we propose a lightweight ontology for wastewater management 2. Wastewater Treatment in by biological methods widely applied in Vietnam. Vietnam
This paper analyzes Vietnam’s biological methods and develops a lightweight ontology for addressing domes- Vietnam’s municipalities [11, 12] generally evolve signifitic wastewater treatment plants (referring to [4]). Here, cantly fast, and constructing urban drainage and wastewwe need an OWL file to be the foundation for building ater treatment is increasingly challenging. Therein, the actual application. We will use description logic to the investment issue in urban infrastructure technidemonstrate the implementation of the code in this paper cal systems in general and urban water collection and because the DLs family [5] provides a solid foundation drainage systems, in particular, needs to be solved. Many of Web Ontology Language (OWL). One of OWL2’s pro- provinces and metropolises have asynchronous infrasifles is OWL2-EL [ 6, 7], dedicated to applications that tructure planning, failing to keep up with urban develuse huge volumes of data where query-answering is the opment. In addition to that, the need for more techmost significant reasoning task. In addition, this paper nical infrastructure for wastewater treatment is quite also focuses on querying wastewater treatment compa- expected. Many inappropriate technological treatment nies and their methods used in that area. In particular, systems have led to untreated urban domestic wastewmost of the knowledge on waste treatment is still be- ater being blatantly discharged into the environment, ing determined and insuficient. The storage method is threatening the ecological environment and becoming a still fragmented and inconsistent, leading to dificulties great challenge for cities in Vietnam. in searching, inferring, and classifying. For the above According to environmental management reports, reasons, implementing the description logic EL⊥ as a household wastewater accounts for 64% of Vietnam’s fundamental core is a suitable and desirable selection. customary wastewater. However, only about 13% of this
In the research process, we decomposed our approach wastewater is treated according to the process. In
Vietinto three prominent stages: (1) the first aspect is to con- nam, 60% of households discharge wastewater into
genstruct an ontology to store WT’s facts which are imple- erally shared sewers. However, most of the wastewater
menting in Vietnam; (
The remainder of this paper is structured as follows: ing its regulatory and enforcement processes rein the next section (Section 2), we give an overview of lated to wastewater treatment. However, there wastewater treatment in Vietnam. Section 3 will pro- are challenges in complying with and monitoring companies classify according to technology groups as follows (the industrial parks and companies are shown in Figure 1): • Traditional technology with biological treatment
by activated mud and biological filter tank; • WT technology with aerobic biological treatment
by adherent growth organisms; • WT technology with long-lasting activated sludge
biological treatment.
Most domestic wastewater in Vietnamese households [15] originates from the kitchen, bathroom (gray water), and black water from the toilet. Black water is usually treated in indoor septic tanks. Gray water is discharged directly into the water system. In several households, black and gray water are released into the soil to fertilize plants. From there, soil pollution and other pollution start to emerge. Generally, most domestic wastewater is treated by the septic method. The flow of Vietnam’s wastewater is shown in Figure 2
wastewater treatment regulations. Some developed countries have robust environmental man- Figure 2: Wastewater flow in Vietnam. agement systems and more straightforward enforcement procedures. 3. Awareness and education: Awareness of wastewater management and treatment in Vietnam is advancing, but there is a need for education and public awareness on the importance of environmental protection and wastewater treatment. Meanwhile, developed countries have a clearer vision of water management and environmental awareness from the public.
Biological wastewater treatment [16, 17, 18, 19] is an essential and integral stage of the wastewater treatment strategy, and it treats wastewater arriving from either residential buildings or industries. They are often called the “Secondary Medicine” process, which removes any contaminants left over after primary treatment. Chemi
Moreover, according to statistics from the General De- cal treatment of wastewater uses chemicals to react with partment of Environment1, 209 industrial parks have pollutants present in the wastewater, whereas biological been operated nationwide, with a total area of 47, 300 treatment uses microorganisms to degrade wastewater hectares. About 80% of industrial parks in Vietnam have contaminants. This treatment relies on bacteria, nemaa centralized wastewater treatment system, and the re- todes, algae, fungi, protozoa, and rotifers to break down maining 20% have not or are investing in a centralized hazardous organic wastes using normal cellular processes wastewater treatment system. Regarding wastewater to stable inorganic forms. This section briefly presents treatment measures, in Vietnam, industrial parks and several popular biological methods used in Vietnam. Classification of biological techniques based on the activity 1https://monre.gov.vn/ of microorganisms is as follows: 3. Biological Methodologies ...
Companies
Kim Lien Truc Bach
Yen So Bac Thang Long (Phu My Hung) (Phu My Hung)
Son Tra
Son Tra
Hoa Cuong Ngu Hanh Son
Cities Ha Noi Ha Noi Ha Noi
Ha Noi Ho Chi Minh
Da Nang Da Nang
Da Nang Binh Hung
Ho Chi Minh Binh Hung Hoa
Ho Chi Minh Canh Doi
Ho Chi Minh Bac Giang
Bac Giang ...
...
Wastewater 2005 2005 2009 2012 2009 2008 2007 2009 2006 2006 2006 2010 ...
Wattage (2/day) 3,700 2,500 7,000 120,000 141,000 30,000 10,000 15,000 15,900 36,418 11,629 8,000 ...
System General General General General General General Private Private General General General General
Aerobic, Anaerotank, Anaerobic (activated mud) Aerobic, Anaerotank, Anaerobic (activated mud) Aerobic, Anaerotank, Anaerobic (with denitrification) Aerobic, Anaerotank, Anaerobic (Reactor tank) Traditional activated sludge Aerobic lake Oxidation ditch Aerobic, Anaerotank, Anaerobic (activated sludge) Anaerobic pond with tarpaulin Anaerobic pond with tarpaulin Anaerobic pond with tarpaulin
Oxidation ditch ...
... 1. The aerobic method [20] is a method that uses • Intracellular destruction: groups of aerobic microorganisms. For their operation, the microorganism must continuously deliver oxygen and maintain a temperature between 20 and 40. This method is applied widely in Vietnam since Vietnam’s temperature is in the range of microorganism activities. It is the decomposition process of organic and inorganic compounds in the presence of oxygen and microorganisms, and at the same time, increasing biomass or intracellular decomposition thanks to a part of organic and energy extraction. The process mechanism goes as follows: • Organic compounds without nitrogen:
+ ( + 4 − 2 )2 −→ 2 + 2 2 • Organic compounds with nitrogen: N + ( + 4 − 2 − 43 )2 −→ 2
+(
2 7 2 + 52 −→ 52 + 22 + 3 2. The anaerobic method [20, 21] decomposes organic substances without oxygen, resulting in methane, carbon dioxide, nitrogen, hydrogen, and others.
() −→ 2 + 2 + 4
+ 4 + 2 + 2 + Microbial cells 3. Anoxic/Anaerotank method: In an environment where oxygen is unavailable, aerobic microorganisms cannot function; instead, anaerotank microorganisms leverage other sources of oxygen.
As a result, the atom and 2− and 3− molecules will appear. Then, the decomposition process produces energy and produces more microorganisms: + 2− | 3− + microorganisms −→
+2 + 2 + more microorganisms
From the popular biological approaches and expert knowledge, we propose a schema of the structure of Biological Wastewater Treatment, presented in Figure 3.
The companies have applied various methods depending on their region. Namely, several companies are involved in A2O technology [22] (a.k.a. AAO) with three manners (anaerobic, anoxic, and oxic). i.e., Kim Lien, Truc Bach, North-Thang Long, some companies utilize SBR paper, we consider EL⊥ concept expressions [25] which [23] (Sequence Batch Reactor) system for the treatment are built according to the following grammar: of biological wastewater containing high organic matter and nitrogen, treatment of wastewater with activated ::= ⊤ | ⊥ | | ⊓ | ∃ . sludge in a fill-and-discharge manner, i.e., Tu Son, Bac where ∈ . Let , ∈ , , ∈ , and ∈ . Ninh, Yen So. Other companies use “Anaerobic Pond with An EL ontology O = ⟨T , A⟩ (a.k.a. knowledge base) lfoat cover”, i.e., Ngu Hanh Son, Son Tra, Hoa Cuong (in comprises two components, the TBox (Terminological Da Nang province). Moreover, because of the soil and Box denoted by T ) and ABox (denoted by A). The TBox type of wastewater in that area, several enterprises have consists of a set of General Concept Inclusion (GCI) axmodified some active ingredients and biological formula- ioms of the form ⊑ , meaning that is more specific tions, i.e., Ho Tay, Cau Nga (they operate the modified than or is subsumed by , and axioms of the form SBR). Hence, to manage the specific information of com- ⊓ ⊑ ⊥, meaning that and are disjoint concepts. panies in diferent regions, we named the instances after ≡ which is a shortcut for ⊑ and ⊑ . The the company name and correction method, i.e., AAO- ABox is a finite set of assertions on individual objects of KimLien, and SBR-HoTay. Although companies have the form () or (, ). mostly stayed the same since the original technology, The semantics is given in terms of interpretations I = our ontology allows using instances of the original name. (ΔI , ·I ), which consist of a non-empty interpretation Regarding the unexplored and innovative approaches, we domain ΔI and an interpretation function ·I that maps have created an “others” concept to store new solutions. each individual ∈ into an element I ∈ ΔI , each concept ∈ into a subset I ⊆ ΔI , and each role 4. Description Logic into a subset I ⊆ ΔI × ΔI .
A summary of the syntax and semantics of EL⊥ is EL is a family of lightweight DLs, which underlies the shown in Table 2. An interpretation I is said to be a Ontology Web Language profile OWL2-EL, considered model of (or satisfies) an axiom Φ in the form of the left one of the main representation formalisms to express column in the table, denoted by I |= Φ, when the correterminological knowledge [24]. The main ingredients sponding condition in the right column is satisfied. For of DLs are individuals, concepts, and roles, which corre- instance, I |= ⊑ if and only if I ⊆ I . Simispond at the semantic level to objects, sets of objects, and larly, I satisfies a concept (resp. role) assertion, denoted binary relations between objects. More formally, let , by I |= () (resp. I |= (, )), if I ∈ I (resp. , be three pairwise disjoint sets where denotes ( I , I ) ∈ I ). An interpretation I is a model of an a set of atomic concepts, denotes a set of atomic rela- ontology O if it satisfies all the axioms and assertions tions (roles), and denotes a set of individuals. In this in O. An ontology is said to be consistent if it has a model.
Otherwise, it is inconsistent. An axiom Φ is entailed by Syntax ⊑ ⊓ of a with no bound variables is denoted as a ground query. In addition, when only contains one atom with no free variables, it is then known as an instance query.
For the , we have I |= if and only if ΦI = ⊤ tahnedK,|= wiitfhafnrdeeovnalyriaifbl∀eIs ⃗|== K(1−,→.. . ,I|=), a.tuFpoler ∃⊥. { ∈ ΔI | ∃ ∈ ΔI . ∅.( , ) ∈ I , ∈ I } of constraints ⃗ = (1, . . . , ) is said to be the certain answer for over K if the ( ⃗) obtained by replacing Table 2 each variables by in (⃗) , evaluates to for Syntax and semantics of E L⊥ every model of K. Therefore, answering can be reduced to answering.
In this paper, we implement the description logic EL⊥ an ontology, denoted by O |= Φ, if Φ is satisfied by every because it can create large ontologies with complexity model of O. We say that is subsumed by w.r.t. an O(n). Furthermore, the constraints of this background are ontology O if O |= ⊑ . Similarly, we say that is largely compatible and appropriate with the knowledge an instance of w.r.t. O if O |= (). An interpretation of wastewater management.
I = (ΔI , ·I ) is said to be fulfilling when each concept In the next section, we will implement this setting name in the ontology is non-empty in I, i.e., for each for our ontology. Specifically, we conduct representaconcept ∈ , ·I ( ) ≠ ∅. tion, reasoning, and query-answering knowledge of wa
The main reasoning task that is considered in termi- ter treatment management in Vietnam.
nological ontologies is classification. It consists in
computing all the entailed subsumptions ( ⊑ ) (and
equivalences ( ≡ )) that hold between atomic concepts of 5. Representation of ViWaT using
iasndoensctroilboegdy,inor[2t4h]e, cwohniccehpfirtsst⊤coonrsi⊥st.sSiuncthraanspfroorcmedinugre DL E L⊥
the ontology into a normal form using a set of rules, and In this section, we present predominantly how to
implethen performing a classification reasoning process us- ment ontology-based modelling of Vietnamese
wastewaing the set of inference (completion) rules (see [24] for ter treatment management using description logic EL⊥
more details). In this paper, we assume that our ontology language. For the presentation, the arisen common
quesis provided in a normal form, to which we apply com- tions is as follows: “Which companies and regions should
pletion rules for classification. This classification step use the Aerobic Lake method?”, What is the treatment
is to normalize the ontology. The reason of conducting method in the Red River Delta? To answer these
quesnormalization is to handle and transform the complex tions, from the knowledge presented in Section 3, our
axioms into the axioms of all atomic concepts to be sim- ontology will focus on the following characteristics: (1)
pler for the representation process. We define the EL⊥ terrain types in Vietnam; (
Definition 1 (Normal Form EL⊥). An EL⊥ TBox is In particular, for (1), we have divided the terrain type
in normal form if all concept inclusions have one of the into three categories: plain, midlands, and hilly. For (
For (4), we divide Vietnamese territory into four regions:
Query-Answering: A query is a first-order logic for- Nord, Central, South, and Highlands. Within each region, mula, denoted = {⃗ | Φ(⃗)} , where ⃗ = {1, . . . , } we have subdivided down to the level of provinces. It are free variables and is the arity of and atoms of Φ(⃗) is presented in Figure 4 (in the right side). We divide are of the form ( ) or ( , ) with ∈ , ∈ each level to facilitate querying information and solvand , are terms, i.e., instances of or variables. ing problems involving overlapping areas and locations. Furthermore, when Φ(⃗) is of the form ∃⃗. (⃗, ⃗) Moreover, government policies will be decentralized to where ⃗ are bound variables called existentially quanti- the provinces; therefore, the distinct groups/levels can ifed variables, and (⃗, ⃗) is a conjunction of atoms store diferent instances that are comfortable for manof the ( ) or ( , ) with ∈ and ∈ , is agement. said to be a conjunctive query (). Note that if we have For (5), based on the structure of the figure 3, we con = 0, is called a Boolean query (). Here, in the case struct a hierarchy of the ViWaT ontology. The concepts will store pieces of information on wastewater treatment methods. Here, we allow the company to coordinate multiple solutions for its wastewater treatment. Since description logic EL⊥ does not allow a “union” concept (⊔) (but only an “intersection” (⊓)), therefore, we cannot represent this situation in DL EL⊥. To solve the above problem, we consider solution combinations as new solutions and store them in the “Creative Method”. Figure 5 represents a hierarchy of WT methods. Furthermore, to keep each method’s originality and ensure no overlap between techniques during the inference process (reasoning), we create a “disjoin” constraint between several concepts, i.e., ⊓ ⊑ ⊥. That is the reason why we used Description logic EL⊥ instead of only using DL EL
In the next sub-sections, we will represent ontology by description logic EL⊥ in the syntactical and semantical side. Specifically, we will present the knowledge in TBox and ABox.
For a knowledge base K B, the TBox T will represent the structure of the ontology. For ease of presentation, we will only express several axioms to focus on answering the common questions mentioned in the section. These axioms will be relevant to ABox and query-answering, which will be presented in the following subsection. Note that we will split the presentation into two parts for TBox: the hierarchy of the ViWaT ontology and the relationship between the two concepts. We assume the “Thing” concept will be a ⊤ concept. The first subsumption structure of ViWaT ontology will be presented as follows:
Moreover, to represent the knowledge between con- Lien company in Hanoi has applied Anaerobic method and cepts, we have several the relations as follows: Binh Hung Hoa company in Vietnam’s South has applied one solution for wastewater treatment.”. The significant ⃗ = {ℎ , , ℎ, , feature is that Kim Lien has used all three wastewater , ℎ } treatment methods (Anaerobic, Aerobic, and Anoxic).
Moreover, all three manners are scattered in diferent We will use the axioms ⊑ ∃. to express the places in the north of Vietnam. We need to determine knowledge. For example, Kim Lien company located in which city Kim Lien company is using Anaerobic. For the HaNoi; we will have a representation as ⊑ second example, we want to ask where Binh Hung Hoa ℎ or ⊑ ℎ company is located in the South and what solution to since Kim Lien Company is a big company so it will be- use to treat wastewater. Based on the TBox in the above long to TopLevel. These rules will enrich the application section, we implement the ABox as follows: implementation. From the inference ability of the description logic, we implement mostly the relationship for the high level (level 0) instead of for the low level (level 3). i.e., ⊑ ∃ℎ .. However, we also implement low levels (level 3) to keep inference independence, i.e., ⊑ ∃ ℎ..
⎧⎪ ⊑ ∃ℎ . ⎪ ⎪ ⎪ ⎪ ⎪⎪ ⊑ ∃ ℎ. ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ⊑ ∃ ℎ. ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ⊑ ∃ ℎ. ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ⊑ ∃ . ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ⊑ ∃ . ⎪ ⎪ ⎪ ⎪ ⎪ T2 = ⎪⎨⎪ ⊑ ∃ℎ . ⎪⎪ ⊑ ∃ℎ . ℎℎ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪⎪. . . ⎪ ⎪ ⎪⎪ ⊑ ∃ℎ . ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎℎ ⊑ ∃ℎ . ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ⊑ ∃ℎ . ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎ ⊑ ∃ℎ . ⎪ ⎪ ⎪ ⎪ ⎪⎪⎪. . .
⎩
Note that, our TBox includes both of parts T1 and T2 to illustrate the knowledge about subsumption and existence rules. Namely, T = T1 ∪ T2. In the next subsection, we will present the facts in ViWaT’s ABox.
This subsection illustrates several notable cases of the great companies in the main cities, i.e., HoChiMinh City, DaNang City, and the Hanoi capital. Therein, our main purpose is for query-answering to verify and evaluate the feasibility of the implementation process, including “Kim ⎧ (1) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ (2) ⎪ ⎪ ⎪ ⎪ ⎪⎪ (ℎ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ (ℎ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ () ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪() ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ () ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ( ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ () ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ( ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎℎ(ℎ ℎ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎℎ( ℎ) A = ⎪⎨ ⎪⎪ (_ ) ⎪ ⎪ ⎪ ⎪ ⎪⎪ − (“_ ”) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ (”_”) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ℎ (, ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎ(, _ ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ℎ (ℎ, ℎ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ℎ (2, ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎ(1, _) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎ(2, _) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎ(3, _) ⎪ ⎪ ⎪ ⎪ ⎪⎪ ℎ(ℎ, _ ) ⎪ ⎪ ⎪ ⎪ ⎪ ⎪⎪. . .
⎩ We allow a company to have multiple instances with the We end this paper with a discussion on queryindexes marked, i.e., KimLien with 1,2, and 3. The fol- answering in the presence of contradictory information. lowing section presents query-answering the questions Indeed, the presence of “disjoin” constraints between conplaced in subsection 5.2 cepts naturally raises the issue of managing conflicting information. There are a large number of methods in 5.3. Query-Answering ViWaT the literature that support query answering even if the knowledge base is inconsistent. Some methods use the Based on available information, we assume that Kim Lien notion of a "repair" which is a maximal consistent subset company is applying the Anaerobic method to answer the of the knowledge base. In general, an inconsistent knowlquestion of where this company is; and the name of Binh edge base admits a large number of repairs, which makes Hung Hoa company to answer the question of where this the task of answering queries computationally dificult company is and what its wastewater treatment is. To (e.g., [26]). This remains true even if we consider that answer the above question, we coded with the following the TBox is stable and reliable and only the assertions of query: the ABox can be questioned (we then talk of assertional repairs). (, ) → () ∧ ℎ (, ) Depending on the used description language, some inconsistency management methods based on assertional ∧ ℎ (, _ ) repairs can be handled in a tractable way. We find for The query means that searching the example the the well-known Intersection of ABox Re _ method applied with and pair (IAR) semantics [27], the so-called the grounded is located at . The answer from ViWaT ontology will repair [28], or the so-called Elect method [29] defined for be = “ 1” and = “ ” partially ordered DL-lite knowledge bases.
The second question’s query will be coded as follows: The Elect method is interesting because it selects a single consistent subbase of the ABox which contains the set (, ) → (ℎ) of so-called accepted or elected assertions. An assertion of an ABox A is said to be elected if for every conflict ∧ℎ (ℎ, ) involving it, there exists at least one other assertion ∧ ℎ (ℎ, ) of such that is strictly more reliable than . When all the assertions of the ABox A have the same level of The second query means searching where BinhHungHoa reliability then the Elect method simply coincides with company is located with variable and what the the one based on IAR-semantics. It is in this sense that method applied with the variable . The result from the Elect method is considered as an extension of IARour ViWaT ontology will be = “ ℎ” and = semantics. “_ ”. It means that BinhHungHoa Work in progress consists of exploring one of the extencompany has used Aerobic Lake to treat the wastewa- sions of the Elect method while maintaining its tractabilter. ity. The idea is to propose an iterative version of Elect which consists in progressively removing from the ABox assertions which are inconsistent with the repair defined by Elect (wrt a TBox T ). Let us denote by Elect(B) the repair obtained using the Elect method on the set of assertions B (and with respect to a given TBox T ). At initial step (step 0), we simply let A0 = A (the initial Abox).
At step i, if there exists an assertion of A / (A ) such that is inconsistent with Elect(A ) (and a TBox T ) then is removed from A . In this case, we let A+ = A − { } and we repeat again this step. The algoFigure 6: The statistic of concepts, axioms and individuals. rithm stops when at step there is no assertion ∈ A which is inconsistent with Elect(A) and the TBox T .
Elect(A) will then be the final repair of the initial ABox
In general, our ontology responds well to the man- A. This iterated process makes it possible to obtain a agement of information about wastewater treatment of better repair than if a single application of Elect was carcompanies. We have also stored more detailed informa- ried out. Note that the computational overhead induced tion, i.e., the performance capacity of the company, the by this extension concerns additional consistency test beginning year, the biological procedure used, and others checks, which are often performed in polynomial time (see Figure 6). in lightweight description logics.
We have laid the first foundations for the development of a lightweight ontology that meets the needs and actual situation in Vietnam to support the management of wastewater treatment of companies. Furthermore, we have provided an application based on ontology-based modeling for Vietnamese wastewater treatment (VWT) through description logic EL to aim at creating the firm foundation for building a semantic web of VWT by OWL2-EL. Using lightweight description logic for representation, reasoning, and querying is discussed in this paper.
The work presented in this paper is one of the initial steps in a long-term efort to create a universal and large repository for a mobile application and website for future works. Moreover, we will also pay attention to handling specific inconsistent cases in ViWaT.
Acknowledgements: This research has received support from the European Union’s Horizon research and innovation programme under the MSCA-SE (Marie Skłodowska-Curie Actions Staf Exchange) grant agreement 101086252; Call: HORIZON-MSCA-2021-SE-01; Project title: STARWARS (STormwAteR and WastewAteR networkS heterogeneous data AI-driven management).