Frame-based knowledge representation

Frame-like structures can be used to represent the facts, objects and concepts. In this context frames must be interpreted as a software stmcture (a frame) in which all characteristics of an object or a concept is described. The simplest example of frame-like structures are the so-called object-attribute-value triplets. Examples of such triplets to describe a chromatographic column are  

Frames can be seen as structures where all relevant information about an object or a concept is collected. As an example the relevant information about a column in a chromatographic method can be represented by a dedicated general frame, COLUMN. Separate columns can be represented by so-called instantiations of this frame. Instantiations are copies of the general frame that contain the characteristics of a specific object, in this example the separate columns. 

COLUMN 1 Tradename Lichrosorb Functionality C8 Column length 30 cm Particle size 4.0 pm Batch number 2347645 Internal diameter 4.6 mm 

COLUMN 2 Tradename p-Bondapak Functionality C18 Column length 30 cm Particle size 5.0 pm Batch number 3459863 Internal diameter 4.0 mm 

In these frames all specific columns that are relevant for the reasoning process of the expert system can be described in a structured and comprehensive way. The frame-based and rule-based knowledge representation are both required to represent expertise in a natural way. Therefore, in most expert systems a combination of rule-based and frame-based knowledge representation is used. The rule base together with the factual and descriptive knowledge by means, of e.g., frames constitute the knowledge base of the expert system. 

---

Knowledge machine (KM) is a frame-based knowledge representation langnage similar to KRL and other KL-ONE representation languages such as Loom and CLASSIC [13-15], In KM, a frame denotes either a class (i.e., type) or an instance (i.e., individual). Frames have slots, or binary predicates, in which the fillers are axioms about the slot s value. These axioms have both declarative and procedural semantics, allowing for procedural inference. 

Instead of representing knowledge in a static way, rule-based systems represent knowledge in terms of rules that lead to conclusions. A simple rule-based system consists of a set of if-then rules, a collection of facts, and an interpreter controlling the application of the rules by the given facts. Other important knowledge representation techniques are frames and semantic networks [1],... 

Knowledge Representation for IQAP. Three knowledge representation schemes, i.e., relational, rule-based and frame-based, and constraint-based, were examined for developing IQAP s knowledge base. For a general description of some knowledge representation schemes (4). 

Demons and Variables. ALEX provides a three-tiered knowledge representation structure of contexts, demons, and variables. Contexts are top-level structures that isolate individual major areas of the problem under analysis. Demons are the central knowledge representation structure they are essentially frame-based information storage nodes with slots both to hold information on the content of the demon and to reflect the impact of one or more variables on the status of the demon. Variables may hold user-input, default, or calculated information. 

Complementary forms of knowledge representation are based on semantic nets and frames (Figure 8.2). Often, they represent just another form to input knowledge. The internal representation is usually based on the predicate calculus. The latter can also be interpreted as a relation of objects ... 

Figure 3 Frame-based representation of knowledge provides a hierarchical format. Like semantic networks the nodes, or frames, can contain multiple pieces of information in slots, and are linked according to defined relationships between two or more objects. An important property of the frame structure is the ability for frames to inherit slot values. Thus, Steel inherits its state value from the Mefa/frame.

At this time, data relative to approximately 220 catalytic experiments for the considered reactions are taken into accoimt in the factual knowledge base. Most of them (60%) concern the alkylation of toluene with methanol to produce xylene, the reaction which is most studied according to the literature. Each experiment is characterized by different parameters stored in frame structures. Such a hierarchical representation allows an easy and rapid retrieval of the information stored in the knowledge base. Particularly, for each experiment corresponding to one particular reaction type, one considers an identification number, the catalyst used with its chemical and physico-chemical characterizations, the reaction conditions, and the results of the catalytic testing, i.e., reactant conversions (conversion), isomer selectivities (selectivity), and product and side product selectivities (productivity) (Fig. 2). 

---