Skrýt

# Differences

This shows you the differences between two versions of the page.

 en:pojmy:din [2019/09/27 10:29]Václav Cvrček [How it works] en:pojmy:din [2019/10/15 20:59] (current)Václav Cvrček Both sides previous revision Previous revision 2019/10/15 20:59 Václav Cvrček 2019/09/27 10:29 Václav Cvrček [How it works] 2019/09/27 10:28 Václav Cvrček 2019/09/27 10:22 Václav Cvrček [How it works] 2019/09/27 10:22 Václav Cvrček [How it works] 2019/09/27 10:16 Václav Cvrček [DIN] 2019/09/27 10:08 Václav Cvrček [DIN] 2019/09/27 10:07 Václav Cvrček 2019/09/27 10:07 Václav Cvrček [DIN] 2019/09/27 10:03 Václav Cvrček created 2019/10/15 20:59 Václav Cvrček 2019/09/27 10:29 Václav Cvrček [How it works] 2019/09/27 10:28 Václav Cvrček 2019/09/27 10:22 Václav Cvrček [How it works] 2019/09/27 10:22 Václav Cvrček [How it works] 2019/09/27 10:16 Václav Cvrček [DIN] 2019/09/27 10:08 Václav Cvrček [DIN] 2019/09/27 10:07 Václav Cvrček 2019/09/27 10:07 Václav Cvrček [DIN] 2019/09/27 10:03 Václav Cvrček created Line 1: Line 1: ====== DIN ====== ====== DIN ====== - DIN (Difference index) is a so called effect-size metric, i.e. a measure designed((see Fidler, M. - Cvrček, V.: {{:​pojmy:​josl-separat.pdf|A Data-Driven Analysis of Reader Viewpoints: Reconstructing the Historical Reader Using Keyword Analysis}})) for the purpose of quantifying the relevance ​of a difference ​between values. DIN is implemented for extracting prominent units from a text (keywords) in the [[en:​manualy:​kwords|KWords]] tool. + The DIN (Difference index) is a so called effect-size metric, i.e. a measure designed((see Fidler, M. - Cvrček, V.: {{:​pojmy:​josl-separat.pdf|A Data-Driven Analysis of Reader Viewpoints: Reconstructing the Historical Reader Using Keyword Analysis}})) for the purpose of quantifying the relevance ​differences ​between values. ​The DIN is implemented for extracting prominent units from a text (keywords) in the [[en:​manualy:​kwords|KWords]] tool. ===== Significance and relevance ===== ===== Significance and relevance ===== - When comparing values (e.g. frequencies of words) we should be aware not only of the statistical significance but also whether the difference under consideration is actually relevant for the description. Statistical significance can be obtained ​by several tests (e.g. chi2 test, Fisher'​s test or log-likelihood test).((It is unimportant ​here that these tests can also be employed as association measures for the extraction of collocations)) Significance is usually expressed as a p-value, i.e. the probability that the difference ​is caused by chance or variation within the data. + When comparing values (e.g. frequencies of words) we should be aware not only of the statistical significance, but also whether the difference under consideration is actually relevant for the description. Statistical significance can be obtained ​through ​several tests (e.g. the chi2 test, Fisher'​s test or log-likelihood test).((It is unimportant ​to note that these tests can also be employed as association measures for the extraction of collocations)) Significance is usually expressed as a p-value, i.e. the probability that the difference ​can be attributed to chance or variation within the data. - Even if the difference is significant ​it does not necessarily ​entails ​that it is also relevant for the description. Even a small difference can be significant when there is a lot of measurements available. That is why the statistical significance information is often combined with the effect-size estimation. + Even if the difference is significant, this does not necessarily ​mean that it is also relevant for the description. Even a small difference can be significant when there is a large number ​of measurements available. That is why statistical significance information is often combined with the effect-size estimation. ===== How it works ===== ===== How it works ===== - In the model example of extracting ​prominent words (keywords) from a text we proceed in the following way. For units which display a statistically significant difference, the **DIN** value is subsequently ​calculated: + In this model example ​for the extraction ​of prominent words (keywords) from a text, let us proceed in the following way. For units which display a statistically significant difference, the **DIN** value is calculated ​in the following way: $$DIN = 100 \times \frac{RelFq(Ttxt) - RelFq(RefC)}{RelFq(Ttxt) + RelFq(RefC)}$$ $$DIN = 100 \times \frac{RelFq(Ttxt) - RelFq(RefC)}{RelFq(Ttxt) + RelFq(RefC)}$$ - where $RelFq(Ttxt)$ is the relative frequency of the phenomenon in the analyzed text (target text) and $RelFq(RefC)$ is the relative frequency of the same phenomenon in the reference corpus. ​ + where //RelFq(Ttxt)// is the relative frequency of the phenomenon in the analyzed text (target text) and //RelFq(RefC)// is the relative frequency of the same phenomenon in the reference corpus. ​ - The formula takes into account the difference between relative frequencies (numerator) in relation to the frequency level of the items under comparison (denominator). This reference frequency level can be represented by their average value, as can be seen in following formula which is equivalent ​with the above (the coefficient changed from 100 to 50 in order to keep the DIN values within the desired range): + The formula takes into account the difference between relative frequencies (numerator) in relation to the frequency level of the items under comparison (denominator). This reference frequency level can be represented by their average value, as can be seen in following formula, which is equivalent ​to the above (the coefficient ​has been changed from 100 to 50 in order to keep the DIN values within the desired range): {{:​pojmy:​vzorecdin2.png?​nolink&​350|}} {{:​pojmy:​vzorecdin2.png?​nolink&​350|}} + ===== DIN Values ===== ===== DIN Values ===== - The DIN values are designed to reach values from -100 to 100, it being understood that: + The DIN is designed to reach values from -100 to 100, it being understood that: * a value of -100 means that the given phenomenon does not occur in the analyzed text and is only in the reference corpus (therefore the word is not prominent in the analyzed text) * a value of -100 means that the given phenomenon does not occur in the analyzed text and is only in the reference corpus (therefore the word is not prominent in the analyzed text) * a value of 0 means that the given phenomenon has approximately the same relative frequency in the analyzed text and in the reference corpus (therefore the word is not prominent in the analyzed text) * a value of 0 means that the given phenomenon has approximately the same relative frequency in the analyzed text and in the reference corpus (therefore the word is not prominent in the analyzed text) - * a value of 100 means that the word occurs only in the analyzed text (it can therefore be a very prominent word ((In such cases it is important to keep in mind that the complete absence of a word from a reference corpus is unusual and special attention should be paid to it; a word might not occur in a reference corpus e.g. because it is very rare, cited from another language etc.))) + * a value of 100 means that the word occurs only in the analyzed text (it can therefore be a very prominent word ((In such cases it is important to keep in mind that the complete absence of a word from a reference corpus is unusual and special attention should be paid to it; a word might not occur in a reference corpus e.g. because it is very rare, cited from another language, etc.))) - In texts of up to 20 thousand words in length and for analysis of [[en:​pojmy:​word|word forms]], DIN values in the 75-100 range can be considered ​to be of interest ​and they indicate that the unit is probably prominent and can be used as a basis for the interpretation of the entire text. + In texts of up to 20 thousand words in length and for the analysis of [[en:​pojmy:​word|word forms]], DIN values in the 75-100 range may be considered of interest, as they indicate that the unit is probably prominent and can be used as a basis for the interpretation of the entire text. - Furthermore,​ the KWords application offers a whole range of additional information for work with keywords. Apart from the list of keywords and their values it is the data dispersion graph (showing the status of the individual keywords in the text), a graph of so-called keyword links, i.e. relations between keywords in the text and also a concordance of keywords ​for an analysis of their immediate ​context. + Furthermore,​ the KWords application offers a whole range of additional information for work with keywords. Apart from the list of keywords and their values, it also features a data dispersion graph (showing the status of the individual keywords in the text), a graph of so-called keyword links, i.e. relations between keywords in the text, and also a concordance of keywords ​which enables the analysis of their immediate ​contexts. The KWords application was also designed for creating analyses of temporal (or other) data series. If the user inputs more texts into the application (the maximum amount is 20), the so-called **multi-analysis** regime is activated. This regime analyzes all the inserted texts and the results of the individual analyses are compared based on the DIN. The KWords application was also designed for creating analyses of temporal (or other) data series. If the user inputs more texts into the application (the maximum amount is 20), the so-called **multi-analysis** regime is activated. This regime analyzes all the inserted texts and the results of the individual analyses are compared based on the DIN.