The most frequently used word by both genders happens to be 'I', which we had expected, nearly all of the characters mention themselves when speaking. However, we noted that the female characters use 'I' 0.12%, which is a small difference, but important none the less. Women characters in Game of Thrones tend to focus their speech on themselves.
Among the most frequently used words by men in the novel are; 'lord', 'he', 'man', 'king', 'father', and 'boy', each of which references a male figure. This concurs with our hypothesis that men's most frequent words will revolve around other men. Interestingly, the most frequent words by women include; 'he', 'lord', 'father', and 'ser'. We had hypothesised that the women's most frequently used words would reflect their roles as mothers, but it seems that they revolve around the male characters instead. Seeing that the novel is based in a medieval-like time period, it is in accordance with the society and culture of the time on which it is based.
We hypothesised that men would use verbs (VB, VBP, VBZ, VBN, VBG, VBD) more frequently than women, and that women would make more frequent use of descriptive parts of speech such as adjectives and adverbs (JJ, RB, WRB). Lastly, we believed that nouns would be used similarly by both genders. As can be seen in our graphs, our hypothesis was not completely accurate.
Women only used adjectives (JJ) and adverbs (RB) 4.13% and 4.34% respectively as their most frequent parts of speech while personal pronouns (PRP) and nouns (NN) accounted for a combined 21.53% of their parts of speech while verbs (VB) only accounted for 4.59%. This may be due to female characters attempts to mimic masculine speech in order to be taken more seriously by male characters. In contrast, men utilized nouns (NN) and personal pronouns (PRP) for a combined 20.8% of their parts of speech, and using verbs (VB and VBP) only 7.97%. Their use of adjectives (JJ) and adverbs (RB) over verbs (8.98% JJ and RB to 7.97% VB and VBP) may signify that male characters are not concerned with asserting their masculinity when speaking.
In general, it appears that most of the House have quite extensive vocabularies, evidenced by the fact that most words consist of less than 1% of a house's speech. The percentages are quite low because the total word count for each House is quite high. For example, the total word count for House Stark is 29,535, yet the count for 'i', their most frequently used word occurs 740 times. Therefore, most of the percentages are surprisingly low. We did not anticipate these low percentages in our original hypothesis. However, House Arryn and House Tyrell are notable exceptions because many of their top 20 words consist of more than 1% of their total speech, which would indicate that these houses have a slightly less expansive vocabulary.
Although 'i' is the most commonly used word for all of the houses, House Tyrell used it the most often. The most houses average at around 2.5-3%, but it accounts for 7% of House Tyrell's total speech. This result was surprising because we anticipated the Lannisters to have the highest use of the word 'i' because of their self-centered nature. However, the members of House Tyrell, in retrospect, are also quite vain, yet they lack the money and prestige of the Lannisters. Instead, House Tyrell is dependent upon physical characteristics, such as strength and beauty.
The most commonly used parts of speech for the house were nouns and personal pronouns. Approximately, half of the houses have nouns as their most commonly used and the other half use personal pronouns most often. Determiners, prepositions, proper nouns, and adjectives are also commonly used by all the houses. Unfortunately, determiners and prepositions are mainly functional words, which is neither surprising nor enlightening. However, adjectives have mid-range use amongst the houses and are most used by House Tarly. Sam Tarly is the only member of House Tarly that is present in the novel and is feminized throughout the novel. His common use of adjectives indicates his feminity because women are often depicted as flowy and descriptive. In this case, Martin uses adjectives to emphasize Tarly's feminity.
As previously mentioned, the Tyrells once again use personal pronouns more often than the Lannisters, who use nouns most frequently. The Lannisters likely use more nouns because they often refer to their wealth. These results would indicate that the Lannisters refer more to their physical possessions than themselves. Since the Tyrells lack wealth, they use personal pronouns more often, when expressing their vanity.
There is only one region that doesn't have "I" as the most common word: the Stormlands. For this book, the characters from this region are Renly and Robert Baratheon; it makes sense that "us" and "brother" would top I. Renly talks mostly about Robert, his brother, and Robert talks mostly about Eddard Stark, a man he identifies as a brother.
The majority of the regions have masculine words as their most frequent, but the Vale has "lady", "mother", and "she". This shows how much a region being run by a woman can influence how the people talk. I would be interesting to see if in later books, when Cersei controls the Crownlands, other regions begin to have similar - and more feminine - most frequent words.
The Reach is the only region that has any part of speech more common than nouns; the most common being personal pronouns. This region is made up primarily of the Tyrells. As noted in the House conclusions, the Tyrells are very self-centered, so it would make sense that their region would reflect their preferences for speech, as they are the largest house.
We found this to be very interesting, since as noted above, we thought the Lannisters, from the Westerlands, would be the most self-centered. But instead, the Tyrells is more self-centered because they have less gold and notoriety to speak of, so they can only refer to themselves.
The most important point we have discovered about the regions, is that they take after whatever the largest house of the area is. The Westerlands speech takes after the Lannisters, the North takes after the Starks, etc.
Out of intellectual curiosity, we created graphs for the most frequently used words by the principle characters in the novel. Through these graphs, users may see how these words are used by characters and how they then factor into word frequency by region, gender, and House.