diff --git a/CHANGELOG.md b/CHANGELOG.md
index 5c57e3373..ea6e06700 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -21,6 +21,7 @@
## [3.6.0](https://github.com/BLKSerene/Wordless/releases/tag/3.6.0) - ??/??/2024
### 🎉 New Features
- Measures: Add effect size - conditional probability / ΔP / squared association ratio
+- Settings: Add Settings - Measures - Effect Size - Mutual Information / Pointwise Mutual Information / Pointwise Mutual Information (Cubic) / Pointwise Mutual Information (Squared)
- Utils: Add Stanza's Sindhi dependency parser
### 📌 Bugfixes
diff --git a/doc/doc.md b/doc/doc.md
index 04127edfa..19810670c 100644
--- a/doc/doc.md
+++ b/doc/doc.md
@@ -1139,7 +1139,7 @@ Readability Formula|Formula|Supported Languages
Al-Heeti's readability formula¹ ([Al-Heeti, 1984, pp. 102, 104, 106](#ref-al-heeti-1984))||**Arabic**
Automated Arabic Readability Index ([Al-Tamimi et al., 2013](#ref-al-tamimi-et-al-2013))||**Arabic**
Automated Readability Index¹ ([Smith & Senter, 1967, p. 8](#ref-smith-senter-1967) Navy: [Kincaid et al., 1975, p. 14](#ref-kincaid-et-al-1975))||All languages
-Bormuth's cloze mean & grade placement ([Bormuth, 1969, pp. 152, 160](#ref-bormuth-1969))| where **C** is the cloze criterion score, whose value could be changed via **Menu Bar → Preferences → Settings → Measures → Readability → Bormuth's Grade Placement → Cloze criterion score**|**English**
+Bormuth's cloze mean & grade placement ([Bormuth, 1969, pp. 152, 160](#ref-bormuth-1969))| where **C** is the cloze criterion score, whose value could be modified via **Menu Bar → Preferences → Settings → Measures → Readability → Bormuth's Grade Placement → Cloze criterion score**|**English**
Coleman-Liau index ([Coleman & Liau, 1975](#ref-coleman-liau-1975))||All languages
Coleman's readability formula¹ ([Liau et al., 1976](#ref-liau-et-al-1976))||All languages²³
Crawford's readability formula ([Crawford, 1985](#ref-crawford-1985))||**Spanish**²
@@ -1153,25 +1153,25 @@ Readability Formula|Formula|Supported Languages
Flesch-Kincaid grade level ([Kincaid et al., 1975, p. 14](#ref-kincaid-et-al-1975))||All languages²
Flesch reading ease¹ ([Flesch, 1948](#ref-flesch-1948) Powers-Sumner-Kearl: [Powers et al., 1958](#ref-powers-et-al-1958) Dutch: [Douma, 1960, p. 453](#ref-douma-1960); [Brouwer, 1963](#ref-brouwer-1963) French: [Kandel & Moles, 1958](#ref-kandel-moles-1958) German: [Amstad, 1978](#ref-amstad-1978) Italian: [Franchina & Vacca, 1986](#ref-franchina-vacca-1986) Russian: [Oborneva, 2006, p. 13](#ref-oborneva-2006) Spanish: [Fernández Huerta, 1959](#ref-fernandez-huerta-1959); [Szigriszt Pazos, 1993, p. 247](#ref-szigrisze-pazos-1993) Ukrainian: [Partiko, 2001](#ref-partiko-2001))||All languages²
Flesch reading ease (Farr-Jenkins-Paterson)¹ ([Farr et al., 1951](#ref-farr-et-al-1951) Powers-Sumner-Kearl: [Powers et al., 1958](#ref-powers-et-al-1958))||All languages²
-FORCAST ([Caylor & Sticht, 1973, p. 3](#ref-caylor-sticht-1973))|
* **One sample of 150 words** would be taken randomly from the text, so the text should be **at least 150 words long**.|All languages²
+FORCAST ([Caylor & Sticht, 1973, p. 3](#ref-caylor-sticht-1973))|
* **A 150-word-long sample** would be taken randomly from the text, so the text should be **at least 150 words long**.|All languages²
Fucks's Stilcharakteristik ([Fucks, 1955](#ref-fucks-1955))||All languages²
GULPEASE ([Lucisano & Emanuela Piemontese, 1988](#ref-lucisano-emanuela-piemontese-1988))||**Italian**
Gunning Fog Index¹ (English: [Gunning, 1968, p. 38](#ref-gunning-1968) Powers-Sumner-Kearl: [Powers et al., 1958](#ref-powers-et-al-1958) Navy: [Kincaid et al., 1975, p. 14](#ref-kincaid-et-al-1975) Polish: [Pisarek, 1969](#ref-pisarek-1969))| where **NumHardWords** is the number of words with 3 or more syllables, except proper nouns and words with 3 syllables ending with *-ed* or *-es*, for **English texts**, and the number of words with 4 or more syllables in their base forms, except proper nouns, for **Polish texts**.|**English & Polish**²
Gutiérrez de Polini's readability formula ([Gutiérrez de Polini, 1972](#ref-gutierrez-de-polini-1972))||**Spanish**
Legibilidad µ ([Muñoz Baquedano, 2006](#ref-munoz-baquedano-2006))| where **LenWordsAvg** is the average word length in letters, and **LenWordsVar** is the variance of word lengths in letters.|**Spanish**
-Lensear Write Formula ([O’Hayre, 1966, p. 8](#ref-o-hayre-1966))| where **NumWords1Syl** is the number of monosyllabic words excluding *the*, *is*, *are*, *was*, *were*.
* **One sample of 100 words** would be taken randomly from the text, and if the text is **shorter than 100 words**, **NumWords1Syl** and **NumSentences** would be multiplied by 100 and then divided by **NumWords**.|**English**²
+Lensear Write Formula ([O’Hayre, 1966, p. 8](#ref-o-hayre-1966))| where **NumWords1Syl** is the number of monosyllabic words excluding *the*, *is*, *are*, *was*, *were*.
* **A 100-word-long sample** would be taken randomly from the text. If the text is **shorter than 100 words**, **NumWords1Syl** and **NumSentences** would be multiplied by 100 and then divided by **NumWords**.|**English**²
Lix ([Björnsson, 1968](#ref-bjornsson-1968))||All languages
Lorge Readability Index¹ ([Lorge, 1944](#ref-lorge-1944) Corrected: [Lorge, 1948](#ref-lorge-1948))||**English**³
-Luong-Nguyen-Dinh's readability formula ([Luong et al., 2018](#ref-luong-et-al-2018))|
* The number of syllables is estimated by tokenizing the text by whitespace and counting the number of tokens excluding punctuation marks|**Vietnamese**
+Luong-Nguyen-Dinh's readability formula ([Luong et al., 2018](#ref-luong-et-al-2018))|
* The number of syllables is estimated by tokenizing the text by whitespace and counting the number of tokens excluding punctuation marks.|**Vietnamese**
McAlpine EFLAW Readability Score ([McAlpine, 2006](#ref-mcalpine-2006))||**English**
neue Wiener Literaturformeln¹ ([Bamberger & Vanecek, 1984, p. 82](#ref-bamberger-vanecek-1984))||**German**²
neue Wiener Sachtextformel¹ ([Bamberger & Vanecek, 1984, pp. 83–84](#ref-bamberger-vanecek-1984))||**German**²
OSMAN ([El-Haj & Rayson, 2016](#ref-elhaj-rayson-2016))| where **NumFaseehWords** is the number of words which have 5 or more syllables and contain ء/ئ/ؤ/ذ/ظ or end with وا/ون.
* The number of syllables in each word is estimated by adding up the number of short syllables and twice the number of long and stress syllables in each word.|**Arabic**
Rix ([Anderson, 1983](#ref-anderson-1983))||All languages
-SMOG Grading ([McLaughlin, 1969](#ref-mclaughlin-1969) German: [Bamberger & Vanecek, 1984, p.78](#ref-bamberger-vanecek-1984))|
* A sample would be constructed using **the first 10 sentences, the last 10 sentences, and the 10 sentences at the middle of the text**, so the text should be **at least 30 sentences long**.|All languages²
-Spache readability formula¹ ([Spache, 1953](#ref-spache-1953) Revised: [Spache, 1974](#ref-spache-1974))|
* **Three samples each of 100 words** would be taken randomly from the text and the results would be averaged out, so the text should be **at least 100 words long**.|English
-Strain Index ([Nathaniel, 2017](#ref-nathaniel-2017))|
* A sample would be constructed using **the first 3 sentences in the text**, so the text should be **at least 3 sentences long**.|All languages²
-Tränkle-Bailer's readability formula¹ ([Tränkle & Bailer, 1984](#ref-trankle-bailer-1984))|
* **One sample of 100 words** would be taken randomly from the text, so the text should be **at least 100 words long**.|All languages³
+SMOG Grading ([McLaughlin, 1969](#ref-mclaughlin-1969) German: [Bamberger & Vanecek, 1984, p.78](#ref-bamberger-vanecek-1984))|
* A sample consisting of **the first 10 sentences, the last 10 sentences, and the 10 sentences at the middle of the text** would be taken from the text, so the text should be **at least 30 sentences long**.|All languages²
+Spache readability formula¹ ([Spache, 1953](#ref-spache-1953) Revised: [Spache, 1974](#ref-spache-1974))|
* **Three 100-word-long samples** would be taken randomly from the text and the results would be averaged out, so the text should be **at least 100 words long**.|English
+Strain Index ([Nathaniel, 2017](#ref-nathaniel-2017))|
* A sample consisting of **the first 3 sentences of the text** would be taken from the text, so the text should be **at least 3 sentences long**.|All languages²
+Tränkle-Bailer's readability formula¹ ([Tränkle & Bailer, 1984](#ref-trankle-bailer-1984))|
* **A 100-word-long sample** would be taken randomly from the text, so the text should be **at least 100 words long**.|All languages³
Tuldava's readability formula ([Tuldava, 1975](#ref-tuldava-1975))||All languages²
Wheeler-Smith's readability formula ([Wheeler & Smith, 1954](#ref-wheeler-smith-1954))| where **NumUnits** is the number of sentence segments ending in periods, question marks, exclamation marks, colons, semicolons, and dashes.|All languages²
@@ -1269,14 +1269,14 @@ Indicator of Lexical Density/Diversity|Formula
--------------------------------------|-------
Brunét's index ([Brunét, 1978](#ref-brunet-1978))|
Corrected TTR ([Carroll, 1964](#ref-carroll-1964))|
-Fisher's Index of Diversity ([Fisher et al., 1943](#ref-fisher-et-al-1943))| where *W*₋₁ is the -1 branch of the [Lambert W function](https://en.wikipedia.org/wiki/Lambert_W_function)
+Fisher's Index of Diversity ([Fisher et al., 1943](#ref-fisher-et-al-1943))| where *W*₋₁ is the -1 branch of the [Lambert W function](https://en.wikipedia.org/wiki/Lambert_W_function).
Herdan's vₘ ([Herdan, 1955](#ref-herdan-1955))|
-HD-D ([McCarthy & Jarvis, 2010](#ref-mccarthy-jarvis-2010))|For detailed calculation procedures, see reference. The sample size could be modified via **Menu Bar → Preferences → Settings → Measures → Lexical Density/Diversity → HD-D → Sample size**.
+HD-D ([McCarthy & Jarvis, 2010](#ref-mccarthy-jarvis-2010))|For detailed calculation procedures, see reference.
The sample size could be modified via **Menu Bar → Preferences → Settings → Measures → Lexical Density/Diversity → HD-D → Sample size**.
Honoré's statistic ([Honoré, 1979](#ref-honore-1979))|
-Lexical density ([Halliday, 1989, p. 64](#ref-halliday-1989))| where **NumContentWords** is the number of content words. By default, all tokens whose universal part-of-speech tags assigned by built-in part-of-speech taggers are ADJ (adjectives), ADV (adverbs), INTJ (interjections), NOUN (nouns), PROPN (proper nouns), NUM (numerals), VERB (verbs), SYM (symbols), or X (others) are categorized as content words. For some built-in part-of-speech taggers, this behavior could be changed via **Menu Bar → Preferences → Settings → Part-of-speech Tagging → Tagsets → Mapping Settings → Content/Function Words**.
+Lexical density ([Halliday, 1989, p. 64](#ref-halliday-1989))| where **NumContentWords** is the number of content words. By default, all tokens whose universal part-of-speech tags assigned by built-in part-of-speech taggers are ADJ (adjectives), ADV (adverbs), INTJ (interjections), NOUN (nouns), PROPN (proper nouns), NUM (numerals), VERB (verbs), SYM (symbols), or X (others) are categorized as content words. For some built-in part-of-speech taggers, this behavior could be modified via **Menu Bar → Preferences → Settings → Part-of-speech Tagging → Tagsets → Mapping Settings → Content/Function Words**.
LogTTR¹ (Herdan: [Herdan, 1960, p. 28](#ref-herdan-1960) Somers: [Somers, 1966](#ref-somers-1966) Rubet: [Dugast, 1979](#ref-dugast-1979) Maas: [Maas, 1972](#ref-maas-1972) Dugast: [Dugast, 1978](#ref-dugast-1978); [Dugast, 1979](#ref-dugast-1979))|
Mean segmental TTR ([Johnson, 1944](#ref-johnson-1944))| where **n** is the number of equal-sized segment, the length of which could be modified via **Menu Bar → Preferences → Settings → Measures → Lexical Density/Diversity → Mean Segmental TTR → Number of tokens in each segment**, **NumTypesSegᵢ** is the number of token types in the **i**-th segment, and **NumTokensSegᵢ** is the number of tokens in the **i**-th segment.
-Measure of textual lexical diversity ([McCarthy, 2005, pp. 95–96, 99–100](#ref-mccarthy-2005); [McCarthy & Jarvis, 2010](#ref-mccarthy-jarvis-2010))|For detailed calculation procedures, see references. The factor size could be modified via **Menu Bar → Preferences → Settings → Measures → Lexical Density/Diversity → Measure of Textual Lexical Diversity → Factor size**.
+Measure of textual lexical diversity ([McCarthy, 2005, pp. 95–96, 99–100](#ref-mccarthy-2005); [McCarthy & Jarvis, 2010](#ref-mccarthy-jarvis-2010))|For detailed calculation procedures, see references.
The factor size could be modified via **Menu Bar → Preferences → Settings → Measures → Lexical Density/Diversity → Measure of Textual Lexical Diversity → Factor size**.
Moving-average TTR ([Covington & McFall, 2010](#ref-covington-mcfall-2010))| where **w** is the window size which could be modified via **Menu Bar → Preferences → Settings → Measures → Lexical Density/Diversity → Moving-average TTR → Window size**, **NumTypesWindowₚ** is the number of token types within the moving window starting at position **p**, and **NumTokensWindowₚ** is the number of tokens within the moving window starting at position **p**.
Popescu-Mačutek-Altmann's B₁/B₂/B₃/B₄/B₅ ([Popescu et al., 2008](#ref-popescu-et-al-2008))|
Popescu's R₁ ([Popescu, 2009, pp. 18, 30, 33](#ref-popescu-2009))|For detailed calculation procedures, see reference.
@@ -1367,9 +1367,9 @@ Measure of Dispersion (Parts-based)|Measure of Adjusted Frequency (Parts-based)|
-----------------------------------|-------------------------------------------|-------
Carroll's D₂ ([Carroll, 1970](#ref-carroll-1970))|Carroll's Uₘ ([Carroll, 1970](#ref-carroll-1970))|
|Engwall's FM ([Engwall, 1974, p. 53](#ref-engwall-1974))| where **R** is the number of sub-sections in which the word appears at least once.
-Gries's DP ([Gries, 2008](#ref-gries-2008); [Lijffijt & Gries, 2012](#ref-lijffijt-gries-2012))||
* Normalization is applied by default, which behavior you could change via **Menu Bar → Preferences → Settings → Measures → Dispersion → Gries's DP → Apply normalization**.
+Gries's DP ([Gries, 2008](#ref-gries-2008); [Lijffijt & Gries, 2012](#ref-lijffijt-gries-2012))||
* Normalization is applied by default, which behavior could be modified via **Menu Bar → Preferences → Settings → Measures → Dispersion → Gries's DP → Apply normalization**.
Juilland's D ([Juilland & Chang-Rodrigues, 1964, p. LIII](#ref-juilland-chang-rodrigues-1964))|Juilland's U ([Juilland & Chang-Rodrigues, 1964, p. LXVIII](#ref-juilland-chang-rodrigues-1964))|
- |Kromer's UR ([Kromer, 2003](#ref-kromer-2003))| where **ψ** is the [digamma function](https://en.wikipedia.org/wiki/Digamma_function), and **C** is the [Euler–Mascheroni constant](https://en.wikipedia.org/wiki/Euler%E2%80%93Mascheroni_constant).
+ |Kromer's UR ([Kromer, 2003](#ref-kromer-2003))| where **ψ** is the [digamma function](https://en.wikipedia.org/wiki/Digamma_function) and **C** is the [Euler–Mascheroni constant](https://en.wikipedia.org/wiki/Euler%E2%80%93Mascheroni_constant).
Lyne's D₃ ([Lyne, 1985](#ref-lyne-1985))||
Rosengren's S ([Rosengren, 1971](#ref-rosengren-1971))|Rosengren's KF ([Rosengren, 1971](#ref-rosengren-1971))|
Zhang's Distributional Consistency ([Zhang, 2004](#ref-zhang-2004))||
@@ -1474,9 +1474,6 @@ Test of Statistical Significance|Measure of Bayes Factor|Formula|Collocation Ext
Conditional probability:
\text{P} = \frac{O_{11}}{O_{x1}} \times 100
-Cubic association ratio:
- \text{IM}^3 = \log_{2} \frac{{O_{11}}^3}{E_{11}}
-
ΔP:
\Delta\text{P} = \frac{O_{11}}{O_{x1}} - \frac{O_{12}}{O_{x2}}
@@ -1508,7 +1505,7 @@ Mutual Expectation:
\text{ME} = O_{11} \times \frac{2 \times O_{11}}{O_{1x} + O_{x1}}
Mutual information:
- \text{MI} = \sum_{i = 1}^n \sum_{j = 1}^n \left(\frac{O_{ij}}{O_{xx}} \times \log_{2} \frac{O_{ij}}{E_{ij}}\right)
+ \text{MI} = \sum_{i = 1}^n \sum_{j = 1}^n \left(\frac{O_{ij}}{O_{xx}} \times \log_{base} \frac{O_{ij}}{E_{ij}}\right)
Odds ratio:
\text{Odds ratio} = \frac{O_{11} \times O_{22}}{O_{12} \times O_{21}}
@@ -1517,14 +1514,17 @@ Odds ratio:
\text{%DIFF} = \frac{\left(\frac{O_{11}}{O_{x1}} - \frac{O_{12}}{O_{x2}}\right) \times 100}{\frac{O_{12}}{O_{x2}}}
Pointwise mutual information:
- \text{PMI} = \log_{2} \frac{O_{11}}{E_{11}}
+ \text{PMI} = \log_{base} \frac{O_{11}}{E_{11}}
+
+Pointwise mutual information (cubic):
+ \text{IM}^3 = \log_{base} \frac{{O_{11}}^3}{E_{11}}
+
+Pointwise mutual information (squared):
+ \text{IM}^2 = \log_{base} \frac{{O_{11}}^2}{E_{11}}
Poisson collocation measure:
\text{sig} = \frac{O_{11} \times (\ln O_{11} - \ln E_{11} - 1)}{\ln O_{xx}}
-Squared association ratio:
- \text{IM}^2 = \log_{2} \frac{{O_{11}}^2}{E_{11}}
-
Squared phi coefficient:
\phi^2 = \frac{(O_{11} \times O_{22} - O_{12} \times O_{21})^2}{O_{1x} \times O_{2x} \times O_{x1} \times O_{x2}}
-->
@@ -1536,18 +1536,18 @@ Measure of Effect Size|Formula|Collocation Extraction|Keyword Extraction
Dice-Sørensen coefficient ([Smadja et al., 1996, p. 8](#ref-smadja-et-al-1996))||✔|✖️
Difference coefficient ([Hofland & Johansson, 1982, p. 14](#ref-hofland-johansson-1982); [Gabrielatos, 2018, p. 236](#ref-gabrielatos-2018))||✖️|✔
Jaccard index ([Dunning, 1998, p. 48](#ref-dunning-1998))||✔|✖️
-Kilgarriff's ratio ([Kilgarriff, 2009](#ref-kilgarriff-2009))| where **α** is the smoothing parameter, whose value could be changed via **Menu Bar → Preferences → Settings → Measures → Effect Size → Kilgarriff's Ratio → Smoothing Parameter**.|✖️|✔
+Kilgarriff's ratio ([Kilgarriff, 2009](#ref-kilgarriff-2009))| where **α** is the smoothing parameter, whose value could be modified via **Menu Bar → Preferences → Settings → Measures → Effect Size → Kilgarriff's Ratio → Smoothing parameter**.|✖️|✔
logDice ([Rychlý, 2008, p. 9](#ref-rychly-2008))||✔|✖️
Log Ratio ([Hardie, 2014](#ref-hardie-2014))||✔|✔
MI.log-f ([Kilgarriff & Tugwell, 2002](#ref-kilgarriff-tugwell-2002); [Lexical Computing Ltd., 2015, p. 4](#ref-lexical-computing-ltd-2015))||✔|✖️
Minimum sensitivity ([Pedersen, 1998](#ref-pedersen-1998))||✔|✖️
Mutual Expectation ([Dias et al., 1999](#ref-dias-et-al-1999))||✔|✖️
-Mutual information ([Dunning, 1998, pp. 49–52](#ref-dunning-1998))||✔|✖️
+Mutual information ([Dunning, 1998, pp. 49–52](#ref-dunning-1998))| where **base** is the base of the logarithm, whose value could be modified via **Menu Bar → Preferences → Settings → Measures → Effect Size → Mutual Information → Base of logarithm**.|✔|✖️
Odds ratio ([Pecina, 2005, p. 15](#ref-pecina-2005), [Pojanapunya & Todd, 2016](#ref-pojanapunya-todd-2016))||✔|✔
%DIFF ([Gabrielatos & Marchi, 2011](#ref-gabrielatos-marchi-2011))||✖️|✔
-Pointwise mutual information ([Church & Hanks, 1990](#ref-church-hanks-1990); [Kilgarriff, 2001, pp. 104–105](#ref-kilgarriff-2001))||✔|✔
-Pointwise mutual information (cubic)**¹** ([Daille, 1994, p. 139](#ref-daille-1994); [Kilgarriff, 2001, p, 99](#ref-kilgarriff-2001))||✔|✔
-Pointwise mutual information (squared)**¹** ([Daille, 1995, p. 21](#ref-daille-1995); [Kilgarriff, 2001, p, 99](#ref-kilgarriff-2001))||✔|✔
+Pointwise mutual information ([Church & Hanks, 1990](#ref-church-hanks-1990); [Kilgarriff, 2001, pp. 104–105](#ref-kilgarriff-2001))| where **base** is the base of the logarithm, whose value could be modified via **Menu Bar → Preferences → Settings → Measures → Effect Size → Pointwise Mutual Information → Base of logarithm**.|✔|✔
+Pointwise mutual information (cubic)**¹** ([Daille, 1994, p. 139](#ref-daille-1994); [Kilgarriff, 2001, p, 99](#ref-kilgarriff-2001))| where **base** is the base of the logarithm, whose value could be modified via **Menu Bar → Preferences → Settings → Measures → Effect Size → Pointwise Mutual Information (Cubic) → Base of logarithm**.|✔|✔
+Pointwise mutual information (squared)**¹** ([Daille, 1995, p. 21](#ref-daille-1995); [Kilgarriff, 2001, p, 99](#ref-kilgarriff-2001))| where **base** is the base of the logarithm, whose value could be modified via **Menu Bar → Preferences → Settings → Measures → Effect Size → Pointwise Mutual Information (Squared) → Base of logarithm**.|✔|✔
Poisson collocation measure ([Quasthoff & Wolff, 2002](#ref-quasthoff-wolff-2002))||✔|✖️
Squared phi coefficient ([Church & Gale, 1991](#ref-church-gale-1991))||✔|✖️
diff --git a/doc/measures/effect_size/im2.svg b/doc/measures/effect_size/im2.svg
index 8cb48ef21..47bfa0cab 100644
--- a/doc/measures/effect_size/im2.svg
+++ b/doc/measures/effect_size/im2.svg
@@ -1,34 +1,41 @@
-