The MALLET tool allows for individuals to manipulate several factors when generating topics, all of which can influence to topics that are produced. The number of iterations, the amount of text, the number of words in each topic, and the number of topics for example, can all influence the outcome. Changing the number of topics affects the topics that the tool outputs by either more or less holistically representing the body of texts that MALLET learns. A smaller number of topics leads to an html output with topics that are very prominent themes within the texts. A larger number of topics leads to an html output with much greater detail and variability based on the texts. When using this topic modeling tool, I recommend using 1000 iterations to ensure that the tool learns as much about the text files as possible, and for number of topics, I recommend inputting 50-100 number of topics to see a great variety of outputs that are still small enough subsets of data to analyze broad themes and topics in the texts.

I was unable to save the metadata for my project files and unable to go back to the classroom to re-do this assignment, so I am unable to identify the percentages of story origins for each topic. However, my three favorite topics for this assignment were “Investigation,” “Murder,” and “Attack.” Investigation came from the list with 75 topics, Murder came from the list with 50 topics and Attack came from 30 topics as well. I speculate that all of these topics involve the majority the Sherlock Holmes’ world of story. Interestingly enough, despite the variations in the lists that the topics come from, there is consistency in theme amongst these topics – an attack, murder and then an investigation is common in Arthur Conan Doyle’s stories.

The amount of metadata provided allows for a variety of questions to surface, such as what the publication dates are for stories involving murder and whether one can identify trends in the frequency of murder in a plot versus an attack, etc.

Using MALLET, the quality of the output is a balancing act. The more topics, the more interpretation there can be. With less topics, it makes it easier to specify one term for topic modeling. With too few topics, however, it becomes a bit uncertain and hard to pinpoint a term. The higher the iteration numbers allows us to work with more precise word combinations. The settings I recommend using for MALLET when topic modeling would be 20 topics/250 iterations/10 words printed. I chose this setting because I felt that when I used a narrower search, MALLET provided me with an easier to read a less complicated set of results. I was able to pinpoint a key term when I was provided with less material to look at.

Three Topics

Hallway/Setting:     Empty House (EMPT)

Angela – room, door, open, window, entered, opened, key, rushed, closed, bedroom, passage, instant, locked, floor, stair, pushed, lock, stairs, led, safe

Mike – door, room, opened, open, heard, key, light, sound, passage, stood, inside, closed, hall, entered, locked, steps, pass, lock, dressing, stair

Office/Text:             The Stockbroker’s Clerk (STOC)

Mike – small, pocket, put, study, drew, cut, papers, attention, eye, safe, examination, bird, piece, left, cigar, mark, thumb, finger, seat, interest

Angela – paper, note, read, letter, letters, book, handed, table, papers, written, message, writing, wrote, address, short, sheet, post, write, importance, document

Attributes:                     Creeping Man Story (CREE)

Mike – face, man, head, hand, dark, black, cried, instant, turned, white, suddenly, figure, opened, quick, sight

Angela – man, face, eyes, dark, figure, looked, tall, head, drawn, black, features, mouth, thin, middle, appearance, deep, huge, beard, nose, lines

The story “Empty House” uses the topic Hallway/Seting the most to describe the appearance and the locations that take place within the story. The story “The Stockbroker’s Clerk” uses the topic Office/Text the least amount of times throughout the story. The topic is used to describe a few locations but other than that it isn’t mentioned that much.

Sample Questions

1. How does the setting and descriptions reflect on the tone of the scene/tone of the overall plot?

1. How does Conan Doyle incorporate description differently than other 19th century authors?

Angela and Mike

After discussing the Mallet project together, we came to the conclusion that less topics makes its easier to identify the subject of the topic. And the more iterations, the better relation of words to the topic. However, too little topics can make the topic modeling too broad and too many iterations will take Mallet longer to come up with the topic models. We recommend 50 topics and 1500 iterations will give the best and fastest results for topic models.

House Decor  (50 topics, 3000 iterations)- house front large round side led window left garden windows small low standing lawn close houses lane high centre building huge gate iron wooden grounds narrow park trees chamber elderly lined walls doors upper ancient rain oak fashioned leads avenue appeared winding entrance century barred modern evidently brick surrounded plainly

This data was taken most from The Speckled Band and least from Resident Patient.

Question 1: When are home landscapes an important part to Sherlock Stories?

Question 2: Can descriptions of the home factor into the plot of the story?

Feminine (100 topics, 200 iterations) lady young woman wife maid child girl left beautiful ill give ferguson making mistress notice devoted possibly nurse poor frances

This data was taken most from A Scandal in Bohemia and least in Six Napoleons.

Question 1: Do women play a prominent role in Sherlock Stories?

Question 2: What can be inferred about the time period from the words used to describe women?

Murder (30 topics, 2000 iterations) found dead body death crime left murder police close attention
finally struck blood part blow remained knife examination stick tragedy

These words were found most in The Final Problem and The Adventure of the Blue Carbuncle

Question 1: Are all of Sherlock mysteries based off of murder?

Question 2: How graphic is the murder described in the stories?

 

 

I used several different numbers and played around with them a bit to find my topics. In all I used 5 different settings (50 topics, 100 iterations, 10 words; 50 topics, 2000 iterations, 10 words; 50 topics, 1000 iterations, 20 words; 100 topics, 2000 iterations, 20 words; 100 topics, 1000 iterations, 20 words) to find my 10 different topics. The topics I choose seemed very similar and gave a basic overview of what the Sherlock Holmes Series could be about. Some words in each general topic were part of a specific story. Still, they also gave insight to what the Sherlock Holmes series is all about.

The more obvious topics that relate to the Sherlock Holmes stories include the topics of Murder, Investigation, Home/Baker Street, and Observation. It makes sense that terms relating to these topics would be found most often in the stories and be grouped together. Holmes investigates murders (as well as other crimes), lives on Baker Street (where he meets with almost all of his clients when they ask him to do investigations for him) and has very advanced tools of observation, which he utilizes to solve the cases he takes on. These 4 topics provide a basic premise of the Sherlock Holmes stories – who is involved, where is it set, and what are the key plots.

The other topics that I choose may not a first glance seem like topics that relate to Sherlock Holmes, they are very important in understanding Sherlock and his methods. Money is a topic that may seen out of place but money is seen as being very important to Sherlock in the stories. Holmes would always like to know in advance how much he will be payed. One can infer that he is so confident in his abilities that he thinks anyone can pay him in advance. Another topic I found was Actions, most interestingly the words “sit” and “laid.” This shows that Sherlock is very comfortable with his surroundings as he listens (another action word) to his new clients stories. A topic that may be seen as sort of a stretch to compare to Sherlock Holmes is Married Life. The terms in this topic start out positive but soon become negative with words like “spite” and “hate.” Sherlock could deal with marriages gone wrong. Clothing (and its descriptors) and Writing can be lumped together almost because they are two things that Sherlock observes the most when he is trying to solve a case. The last topic, Time Measures included words “time,” “week,” and “days” so the reader can tell that even though Sherlock is a great detective, it may take him some time to figure out the cases that he takes on.

The MALLET tool was easy to use and allowed me to change the settings to get the results I wanted as much as I wanted. I believe the topics that I found, even the ones that don’t seem obviously connected to the Sherlock Holmes stories, provide a background of information for someone trying to get the general themes and ideas behind the stories, whether they be Digital Humanities students like us or just first-time readers.

50 topics 

Feminine

5.	lady young woman wife maid child girl left beautiful ill give ferguson making mistress notice devoted possibly nurse poor frances

75 Topics

Sherlock’s Mannerisms

61.	holmes chair sat fire pipe back asked arm rose glanced

Decor

67.	light dark side lamp low match floor wall darkness held

100 Topics

Detective

40.	london criminal learned dangerous attempt career order begin failed prepared chance due care fair succeeded

Travel

44.	train station carriage line reached journey bridge hurried drove roof observed body opposite walk town

Murder

62.	body lay blood shot found revolver blow dead head knife drawn fell weapon stick picked

Door

81.	door room opened key safe locked side inside closed lock open shut fastened conscious doors

Assistant

88.	watson dear surely fellow complete touch dangerous hope meet read draw report final form sufficient

500 Words

Setting

40.	front house high drive road trees standing park windows place miles building sun trap drove

Furniture

178.	bed bell rope end ventilator cord mantelpiece hung observe chair pull inches ring dummy fastened

Mallet is a useful tool specifically for looking at the broader idea of an author or a story. It definitely is a distant reading tool, as I could not see it being of value to anyone who is close reading. Mallet is very helpful in that it takes minutes to go through all the information (in this case, every Sherlock Holmes story) and sort it all into groups based on a similarity. This organization would take a person extended periods of time to just read through and record all the words in the Sherlock Holmes stories, much less sort them all into hundreds of groups based on patterns.I thought it was particularly interesting what the tool said about the collection of Sherlock Holmes stories.
The grouping of the words was rather obscure and more often than not, very difficult to figure out. Even when I had 500 groups, it was not as simple to understand the groupings as it would be if a person sorted the words. Had a human being grouped the words, I think it would have been much easier to figure out the pattern of the groupings. Often times, I found that if a group of words was rather obvious, there would be one or two words in the group that came almost out of left field completely. This was the most difficult or frustrating part about using Mallet.

The words cloud really emphasizes the mystery and action of a Sherlock Holmes story. Some of the biggest words are danger, fasten, drove, and body. Danger is one of the key elements that adds to the mystery in any Sherlock Holmes story. Drove and fasten can go together well. Fasten often brings to mind the thought of fastening your seatbelt when on a dangerous or intense ride. In a lot of stories, there is a murder or a body to be found, so that makes sense as to why that is a big word. These words together properly give the essence of a good Sherlock Holmes story.