The results of my topic modeling graphs were incredibly interesting to me. In my first analysis, I combined the topics of “crime,” “police work,” and “murder/death”. In this graph I found that, while the topic of police work tends to spike at random points throughout 1903-1904, crime and murder/death tends to be lower and typically the same throughout the two years in question. I did a bit of research and noticed that the two years graphed in this analysis were years of important sports events and a few protests. This would explain the spike in police activity without a correlating spike in crime.

The second graph I charted was comparing the topics of travel and time. Travel appears to spike in a dramatic increase in 1908, with time spiking upwards as well during this time. I discovered through outside research that throughout the year of 1908 travel was becoming increasingly popular: the year beings with two expeditions around the world (one specifically from New Zealand to Antarctica); the Olympics were held in London in 1908 (which would increase travel to the area); and finally, the first aircraft manufacturing company in England is found in London. This would explain the increase in travel, as well as time.

Another set of topics I compared and contrasted were Business/Commerical and Construction. I noticed in my graphs that there was a correlating increase in both topics in 9125 and I wondered why that was. Looking at outside research, I noted that in 1925 there was a great deal of economic/commercial events taking place. For instance, primogeniture (or the rule that the first born son would inherit from the father) was abolished, Britain returned to the gold standard, the government granted a subsidy to the coal industry while they investigated its issues, the first double-decker buses with covered tops were introduced, and various bridges and tunnels were constructed. These events would clearly influence the Sherlock Holmes stories, as well as explain the increasing spike of these topics in 1925.

Lastly, I decided to compare the topics of literature, description of clothing, and emotional verbs/actions. I thought that these topics were comparable as they all had to do with writing, and to an extent, education.I found the most interesting time period in this graph to look at was the years of 1891-1893. I found that literature often spiked dramatically first, and then the descriptive words would follow. One interesting fact that I discovered was that in 1891 Elementary Education was made free, allowing for an increase in literacy and education. This would, therefore, explain why the spikes were so dramatic around this time period. Futhermore, in 1892 Scottish universities began accepting women, and in 1893 the Brontë Society is established (the oldest literary society) and the Elementary Education Act raises the age to leave school to 11. All these historical events were significant in the rise of literacy and education, therefore explaining the rise of the topics of literature and the following descriptive words and actions.

 

The first topic is Travel:

 

In this graph, we see an increase in travel around 1893 and the only other spike that occurs is later on in 1904, but the 1904 spike is not as high as the spike in 1893, therefore, I decided to research why that might have happened.  I found out that by the end of the 19th century, they invented a new method of transportation.  Based on the website Primary Homework Help The Victorians , “In the 1890s they could travel by motor car.”  Based on the research, I think that people decided to travel more after the invention of the motor car which explains the spike in 1893.

The second and third topics are Writing with Business:

In this graph, I decided to compare the topics writing and business.  These topics both seem to have a spike at about the same time; Writing in 1903 and Business in 1904.  Therefore, I decided to research this further to find out why this might be.  The amount of writing words appear the most in “The Adventure of the Three Students”.  After reading the plot on the Wikipedia article, there is a lot of writing going on in the story because of the fact that it deals with students and a university.  However, it does not explain why business words showed up often, therefore, I looked at another story that was published in 1904.  Based on the Wikipedia article, business words appear pretty frequently in the story “The Adventure of the Abbey Grange” because it talks about how a man has been killed by the Randall gang.  It is interesting why these words tend to rise and fall together; it helps us understand the stories better because it will explain that the stories’ topics will be about writing or business.

The fourth topic is Detective Case:

In this graph, we see a spike, that is higher than the other peak, in detective cases around 1891.  Then, I decided to research why this spike happened when it did.  Based on the Wikipedia article about the Whitechapel Murders, “The Whitechapel murders were committed in or near the impoverished Whitechapel district in the East End of London between 3 April 1888 and 13 February 1891.”  Based on this research, It is possible that the Jack the Ripper case influenced the amount of detecting words in the Holmes stories in 1891.

The fifth topic is Death:

In this graph, we see a spike around 1903 regarding death and there is no other spike like that one throughout the rest of the graph.  Based on the website The Guardian, “During the 1880s and 1890s, local authorities, the LCC and the Metropolitan Public Gardens, Boulevard and Playground Association began to clean up and reopen old burial sites.”  It is possible that the actions of these authorities influenced the amount of death words in the Holmes stories based on the fact that from 1893 onward there is a steady rise in the amount of death words.  However, after the researching, I still am not able to explain the sudden peak in 1903.

The sixth and seventh topics are Time with Crime:

In this graph, we see a spike for both time and crime in the year 1904.  Based on my research in the Wikipedia article of the story “The Adventure of Charles Augustus Milverton”, which was published in 1904, it is about the crime of blackmailing.  It explains how in order to help solve the case, Holmes visits Milverton’s Hampstead house, disguised as a plumber, in order to learn the plan of the house and Milverton’s daily routine.”  Therefore, daily routine refers to time.  Even though, it is evident that crime and time words appear in every Sherlock Holmes story.

The eight and ninth topics are Physical Description with Building:

I decided to pair these two topics together because I wanted to see if there is a correlation between the two and also because they are both descriptions.  Based on this graph, the amount of physical descriptive words and building words tend to rise and fall together.  Except in the year 1904, the amount of building words increases and the amount of physical description words is not as high.  Then, after 1905 they do the complete opposite of each other; when the amount of building words rise, the amount of psychical description words fall or vice versa.  It’s possible that this kind of correlation tells us  that either the story will have more building words or that the story will have more physical description words.

The tenth topic is Emotion:

In this graph, we see that there is a spike in the following years where emotion words show up most frequently, 1893, 1904, 1913, and 1924.  I have come to the conclusion that the stories that were published in these years all contained woman characters, based on the Wikipedia articles, “The Adventure of the Cardboard Box”, “The Adventure of Charles Augustus Milverton”, “The Adventure of the Sussex Vampire”, and the bubble news article.  Based on the fact that they all contained woman characters, It’s possible that the amount of emotion words increased during these times because in Victorian times women were not considered equal based on the Wikipedia article.  This helps us understand the stories better because we can connect them to how the past really was.

100 topics, 15  topic words, 5,000 iterations

Face/head: eyes face lips pale expression hands eager hot thin white cheeks fixed grey brow emotion

Realtionship: life woman love heart knew loved break mine truth evil power world women hands told

Money: hundred money ten business pounds year thousand worth large made pay paid price fifty sum

Detective: facts case theory simple fact points difficulty formed suggest correct impossible idea test remained explanation

Travel: train station carriage line reached journey started bridge drove return hurried cross roof passing save

Protection/Security: door room key opened safe closed inside long study lock locked shut entered bag fastened

Room Descriptions: round corner side table front left stood dressing covered top square carpet gown furnished books

Appearance: man face dark tall hair features eyes head middle thin figure clean beard cut gray

Writing: note letter paper book letters wrote written writing write read handed date envelope slip address

Time: years ago time twenty months thirty week age daughter weeks lives quiet meet retired engaged