Poetry: the basics of algorithmic interpretation Part 2


Then X will appear in the form of the final AMKL in interval form, the conclusions To their frequencies (in this case, estimates of G) will correspond to the peculiarities of the versification of certain artists or groups. For a rough visual representation of the results obtained as a rhythmic (wave) process, we can construct the following diagram in the form of narrow rectangles. Take a standard entry AMCL a few insights To with the highest estimates of G, one for Z = 1 and another for Z = 0. The ordinate going to get the same for a given positive values For G ("amplitude waves To States of the object") and lower negative G (Z = 0). On the abscissa going to get the line numbers (or points in time t) of X that correspond to the selected K.

Thus, the origin (0, 0) would correspond to a rectangle with a maximum height of G (for example, for Z = 1) for the first output and To most of the "old" line of researched poems To do this (in the future, for other To the zero abscissa may be shifted to the left). The same rectangles (there will be D) will be located to the right through some time intervals. Similarly will be located below the rectangles for another Z already with their most equal of the gaps between the rectangles correspond to the absence of the selected situations To during this time. Note that all these rectangles can complex way to alternate all depends on the moments of t for each of them will correspond to their values Z. This diagram will clearly show the most frequent variant of poetry of the author.

You can continue further all of this sequence of builds for all other pairs K, ordered according to their G (and in the end it is possible for the remaining single). Through the middle vertices of all rectangles (many of them will have different height D) is also possible to perform a smooth undulating curve approximately showing the features of the rhythm of the poetic text of the author or their groups. This chart is interesting because the line numbers on the abscissa allow you to learn the context of some selected researcher of the lines of the poems included in It to further information by searching other sources to try in more detail to interpret (understand) the additional features of some situations K. Thus, line numbers, "covered" To play the role of hyperlinks, extending the meaning of the verses.

3. "Text grid. The options of the text grid"

Temporal grid of the verses corresponds to the list of line numbers of poems by each calculated To the entire logical model AMCL (private example is the interpretation of such grid, see above at p. 2). The presence of all other nets (local, personal, reference, modality) is determined by the occurrence of the corresponding variables in each output K.

4. "Harmony and symmetry in poetic text"

Implementation of algorithm AMCL starts by splitting the objective function Y at two levels of values (usually the median), this function has a binary form Z = 1 or Z = 0, i.e. two classes of function values Y, to complement each other. It is also possible to use multivalued logic Z = 0, 1, 2. General, that is the manifestation of symmetry as the equivalence for the researcher allocated some of the state groups of the studied object (for binary case).

Display of poetic texts in the form of a logic model reveals their special order, harmony, the successful alignment of the real informational structure of creative consciousness (and subconscious) perceived as beautiful. Note that this harmonization is essentially a recording of a very extensive source of external information in a more limited and compact information (model) as it is in the form of hyperlinks, allowing readers to quickly connect to new sources of information may already exist in their minds. Here see also p. 2.

5. "Externally, any work tightly (no addenda impossible), internally a truly perfect work openly to build new meanings"

Build AMCL completed the calculation of its dead-end disjunctive form (definition of the minimum number To big G, necessary and sufficient to "cover" all rows from X). Adding any rejected It does not make sense to implement the model as a General external structure is sufficient for recognition of any row of H. Here, the meaning of each of the corresponding output is only set To its variables.