The Application of Prolog to the Development of QA and DBM Systems

Futó, Iván; Darvas, Ferenç; Szeredi, Péter

doi:10.1007/978-1-4684-3384-5_13

Cited by 22 publications

(10 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Using HCs for this purpose offers the advantage that an element of a HC can also be interpreted as a graph element. This possibility (utilized advantageously earlier, Futo et al 1979; Chang and Lee 1973) was also exploited here at developing METABOLEXPERT.…”

Section: Model and General Axioms Of Metabolexpertmentioning

confidence: 97%

“…In this case, the model is expressed by using a kind of mathematical logic, the first-order predicate calculus (FOPC) (Kowalski 1979;Chang and Lee 1973). Application of the model to a particular problem such as finding a structure-activity relationship or predicting drug interactions (Futo et al 1979) means the definition of the general axioms (GA) of a first-order theory (FOT), while application of the problem to a particular compound (or compound set) requires the definition of specific axioms (SA) of the problem in question. A set of SA models a problem in the same concise manner as a QSAR equation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Metabolexpert: An Expert System for Predicting Metabolism of Substances

Darvas¹

1987

QSAR in Environmental Toxicology - II

View full text Add to dashboard Cite

An expert system for simulation of the metabolic fate of substances in humans is presented and exemplified by metabolism of chloramphenicol analogues. The system uses first-order logic and automatic theorem-prover to infer qualitative biological data, thus complementing the quantitative world of QSAR. It uses empirical knowledge of chemists involved in metabolism research. By giving estimates of the expected metabolic routes and metabolites of an administered subtance, METABOLEXPERT will help to orient researchers whO are not expert in drug metabolism. METABOLEXPERT runs on an IBM AT with 640 Kbyte core memory, a 10 Mbyte hard disk, graphics card and colour monitor.

show abstract

Section: Model and General Axioms Of Metabolexpertmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

Metabolexpert: An Expert System for Predicting Metabolism of Substances

Darvas¹

1987

QSAR in Environmental Toxicology - II

View full text Add to dashboard Cite

show abstract

“…Rather than using a Turing-complete language, these approaches respectively use Datalog [2] and QL [8], languages that only offer a subset of the PROLOG language by e.g. limiting the possible forms of recursion and excluding the definition of data structures.…”

Section: Related Workmentioning

confidence: 99%

Time warp, an approach for reasoning over system histories

Gómez

Kellens

Brichau

et al. 2009

Proceedings of the Joint International and Annual ERCIM Workshops on Principles of Software Evolution (IWPSE) and Software Evol

View full text Add to dashboard Cite

The version history of a software system contains a wealth of information that can assist developers in their daily implementation and maintenance tasks. By reasoning over the role of certain code entities in previous versions of the system, developers can better understand their current state, assess the required maintenance and avoid making the same mistakes over and over again. Unfortunately, current approaches do not offer a means to easily extract specific information about the source code from such a version history. In this paper we present Time Warp, a library of logic predicates that builds on the SOUL language and the FAMIX and Hismo meta-models and that allows writing queries about the history of a system. By means of a number of concrete examples, we demonstrate how our approach can be used to express interesting queries over the version history of a system.

show abstract

“…It is less widely known that Prolog a lso offers several di stinct adva ntages as a database system. Although there are currently database sys tems which use logic-based lang uages [ 13,25], these systems a re ge nerally intend ed fo r specialpurpose appli ca ti o ns. particularly in the areas of Artificial Intelligence resea rch.…”

Section: Advantages Of Eachmentioning

confidence: 99%

Simplified integration of Prolog with RDBMS

Berghel¹

1985

SIGMIS Database

View full text Add to dashboard Cite

In this paper we describe several adv(llllllges of integrating rhe declarative. logic-based language, Prolog, with conventional re/arional database managem ellf .\yslems. We thell explain a file conversion procedure which can be used 10 exchange in/ormation belu'een systems. 771e result is a research environment in u'hieh (h e liser can derive maximum benefit from bOlh Prolog and relational database systems.S o rne preliminary defini ti ons a rc in order before we proceed. First, a relati o nal database management system (hereafte r, RDBMS) is a database sys tem based up o n the data model of Codd [5]. On thi s accou nt , a ll database files must be represe nted as rectangular tables. where the columns correspond to attributes and the rows, tuples. Thus. we say that a given relati on, R. co nsists of tup les t t, t2 .... t n defined with respect to a ttributes al,3 2 .... , am' Tuples in a Codd re la tion th ereby corres po nd to records and attributes to field s, in th e pa rla nce of traditio na l file management sys tems.Further. there are so me addit io nal requ irements of legitimate relat ions. Fi rst, relationa l fi les must be 'Oat'. That is. eac h tuple must contain the same number of va lues, and be defined ove r th e sa me se t of a ltri b utes. In addit io n, a ll relati ons must have so me su bset of attribu tes which unique ly define a ll tuples (t ha t is, they must have a primary key). Lastly, re latio ns must be ' normali zed' in orde r to avoid maintenance anoma lies. Whi le relational normali zat ion is defined in terms of six levels [5, 9. 10] we need only conce rn o urselves with the fi rst three. alOm ic formula , 'Pj', where P stands for the pred icate 'is ta ll' whilej stands for the s ubject constant, J o hn. In addition to predica tes and consta nts, predicate logic a lso conta in s th e quantifi ers ' th ere exists' (3) and 'fo r a ll' (V) a nd a denumerable set of quantified va riab les designated by lower-case. final lellers of the Roman a lphabet with or with o ut sub-and super-scripts.The recursive definition of well-formed predicate logic expressions is thus: H. L. Berghel is with the Department o/Computer Science. University 0/ Nebraska. We wi ll say that an a ttribute, B, is/ullctionallydependem upo n a not he r a uribu te, A, if and on ly if for each tuple in the relation, th e va lu e of A uniquely determines the va lu e for 8. For example, a database may co nt ain Social Security numbers and names. Since we can always determine the lauer from the former, we say that names a re functionally depend ent upo n Socia l Secu rit y numbers.A transilive/un crional dependency of C up o n A ex ists if and o nly if C is functionally dependent upon Band B is functionally dependen t upon A, for a tt ributes A, Band C. Further, C is!II11r!lInclionoll,, dependenl upon A and B if and o nly if C is functio nally dependent upon their combination, but no t up o n ei ther, indi vid ually.We will rest rict o ur discuss ion to relations in Third Normal Form (3NF) [5]. A rela ti o n is in 3NF if and...

show abstract

The Application of Prolog to the Development of QA and DBM Systems

Cited by 22 publications

References 4 publications

Metabolexpert: An Expert System for Predicting Metabolism of Substances

Metabolexpert: An Expert System for Predicting Metabolism of Substances

Time warp, an approach for reasoning over system histories

Simplified integration of Prolog with RDBMS

Contact Info

Product

Resources

About