**INDUCTIVE CONCLUSION** — type of conclusions from group of isolated facts to a nek-swarm to the hypothesis «generated» by these facts.

Usually the hypothesis has an appearance of the general statement and is called inductive generalization. And. century shortly call induction. Induction is exactly conclusion, but not a conclusion as no formal rule directing the train of thought from the facts parcels to a hypothesis investigation can be specified. But the basis for reasonable confidence in the validity of a hypothesis nevertheless is available — the made observations and experience confirm it. Thus, essence And. consists in «anticipation» and in «anticipation» on the basis of last experience, in the movement of a thought advancing the facts century. Medical diagnosis (see) in essence represents And. century as in most cases the doctor has no sufficient information for absolute conviction in correctness of the made diagnosis. Moreover, in considerable number of cases when identification of an absolute symptom of this or that disease is possible (e.g., bacterial, or serol, confirmation of an infectious disease), it demands considerable time. In these conditions the doctor is forced by comparison available a wedge, manifestations and the information about the patient with personal experience to make the decision on the most probable diagnosis and on the basis of this decision to appoint the corresponding treatment.

It is natural that even in the presence of very weighty and often repeating in a combination with any phenomenon facts the generalizing hypothesis can be wrong. Thus, And. it is comparable to decision-making in the conditions of uncertainty century. At conclusion on induction consciously or unconsciously are guided by assessment of probability of the conclusion. The inductive probability is degree of a justification of the conclusion parcels, a measure of justice of transition from the facts to an inductive hypothesis. This measure can be intuitive, everyday, based on value judgment, but it can be defined also by more exact criteria. If this measure is rather high, it generates in us reasonable confidence in the validity of inductive hypotheses.

Universal cataloguing And. is not present century. Allocate usually full induction when generalization belongs to final and foreseeable area of the facts and all of them enter sending of induction, and incomplete induction when generalization belongs to infinite or final and vast area of the facts. Incomplete induction is subdivided on logical and statistical. The last is applied where the sampling method of observation with explicit assessment of probabilities of distribution of measurable properties of bodies of interest is possible about what preliminary hypotheses become. Statistical induction consists in acceptance or a deviation of each such hypothesis on the basis of the schemes of a conclusion providing different estimates of the made hypotheses. There are no standard ways of justification of logical induction yet as they are absent and for statistical schemes which come true only the fact that seldom yield wrong results. If experience allows to formulate several alternative hypotheses, logical induction can be reduced to the deductive scheme of a dividing indirect demonstration.

Other way of justification is connected with explicit assessment of probability And. century by means of calculation of so-called function of confirmation of a hypothesis the facts of observation.

**Bibliography:** Logic and empirical knowledge, under the editorship of P. V. Tavanets, M., 1972; Induction, acceptance and rational belief, ed. by M. Swain, Dordrecht, 1970.

*M. M. Novoselov.*