HathiTrust+Bookworm (HT+BW) visualizes word trends in 13.7 million works held by the HathiTrust. It is currently being developed under a two-year NEH Implementation Grant, but it can already be tried at https://analytics.hathitrust.org/bookworm.
The world's great research libraries have, over time, carefully assembled a rich body of metadata pertaining to the books in their collections. Since the HTRC has access to volume-level metadata as well as volume-level content, we have constructed a Bookworm of the HathiTrust public domain corpus. We felt that setting up Bookworm with a HathiTrust corpus would provides scholarly researchers with the means of exploring trends.
This tool enables scholars to discover new textual use patterns across the entire corpus. In the future, we plan to ingest the entire HathiTrust corpus and continue to identify appropriate metadata to use for the faceted browsing. This tool will be particularly useful to scholars interested in books that are still under copyright - which is the case for most books published after 1923. Although these books will not be available for reading or downloading online, working with individual words and phrases and tracking their occurrences through time will be useful to academic researchers, especially historians, sociologists and literary scholars.
John Unsworth has noted that a fundamental goal of the humanities is appreciation: "by paying attention to an object of interest, we can explore it, find new dimensions within it, notice things about it that have never been noticed before, and increase its value" (2004). Shifting from traditional close reading to a large-scale view of text presents a profound discomfort for humanities scholars, due to the difficulty in retaining the same sensitivity to what is actually contained in the works being studied. HTRC-Bookworm will function as a link between quantitative analysis (distant reading) and close reading. According to Frederick Gibbs and Daniel Cohen, "any robust digital research methodology must allow the scholar to move easily between distant and close reading, between the bird's eye view and the ground level of the texts themselves" (2011). This is what HTRC-Bookworm intends to accomplish (within the limitations of applicable copyright laws.)
Soda vs Pop
Soda or Pop? How do Americans in different places refer to their soft drinks? Besides all kinds of scientific papers and journal articles arguing about it, the Pop vs Soda (http://popvssoda.com/) , a web-based project, plots the regional variations in the use of the terms "Pop" and "Soda" to describe soft drinks. Current statistics are available here at http://popvssoda.com/statistics/USA.html. Their statistics and mappings are interesting to read.However, what if we want to look back into the history and find the hidden statistics? Where can we get historical evidence for our question? What will the results look like if the statistics are based on authorized publications rather than people's voting online? Try Bookworm! With data extracted from millions of publications from 1940 to 2015, we visualized the” Soda to Pop Ratio” by State .Y-ray represents the publication states while X-ray shows the word ratio of “soda” to “Pop”. For example, we can see from the graph that publications in Massachusetts use Soda for the soft drinks almost ten times the frequency of using Pop.
Soda to Pop Ratio Graph made with Bookworm Data
We also visualize the Soda to Pop ratio with the statistics from Pop vs Soda Page. Now you can look at a whole picture to find answers to more questions. What are the states where the word “Soda” always dominate in publications? Are publication language sharing the same words preferences for soft drinks as people’s oral language do? Start your exploration with Bookworm!
Soda to Pop Ratio Graph made with the Pop vs Soda Page statistics
Peking to Beijing
What is the capital city of China? Interestingly, some would say Peking, while the others would say Beijing. Referring to the same city, Beijing is pretty close phonetically to the original Mandarin while Peking has been used for a longer time internationally. Some findings argue that the Chinese government is insisting on the more modern transliteration Beijing rather than Peking. What’s more, they claim that with China’s rapid development and increasing power, the trend of replacing Peking with Beijing grows. To further investigate this argument, we used Bookworm to find out the word usage in publications of six countries from 1960s to 2010s. Then we generated a graph showing the log ratio of Peking to Beijing grouped by Country. Y-ray marks the publication country while X-ray shows the time of the publications. Blocks of different colors indicate different ranges of the ratio. Click on a block and you will find a list of related publications during a certain period in the country you pick. Try different settings and input various words, you will find more!
Log Ratio of Peking to Beijing Usage, By Country
Part of the lists of related publications
Inuit or Eskimo
Another typical word preference discussion is “Inuit or Eskimo”. The term "Eskimo" is commonly used in Alaska to refer to all Inuit people, however, this name given by non-Inuit people is considered offensive in many other places. Linda Lanz, a Ph.D. in linguistics from Rice University in Houston, claims that “In Canada, the term Inuit is preferred over Eskimo, which is considered offensive.” She sums it up with “Canada: Inuit; United States (i.e., Alaska): Eskimo.”
Doctor Lanz is not the only one who thinks so. Actually, her argument is in line with many linguistic studies. But “common opinions” are not necessarily the truth. To verify such theory, it is helpful to visualize word popularity of Inuit and Eskimo with Bookworm. According to the Bookworm output, for United States, Linda’s argument definitely makes sense in the period from 1880 to 1980. However, since 2000, the situation has changed. To further examine such change, we use the “Dates” settings on the panel to “zoom” in. Then we can see more details. For example, we can have a close look at the turning point when word counts of Inuit first surpassed that of Eskimo in the United States.
Word Popularity lines of Inuit VS Eskimo in United States and Canada from 1760 to 2000
Word Popularity lines of Inuit VS Eskimo in United States and Canada from 1981 to 2010
The word count lines of Canada is also dramatic, challenging the “common opinions” on Inuit vs Eskimo issue. As we zoom in, we can tell that for some reasons, the use of Inuit surged from 1968-1984 while the use of Eskimo dropped quickly at the same time. Why? What caused such changeover? Was it the same case in Canadian people‘s oral speaking? With such findings, we can raise more questions against the previous arguments and start our own research with new evidence.
Word Popularity lines of Inuit VS Eskimo in United States and Canada from 1957 to 1989
Popularity of years, By year
Bookworm gives users a simple visualization platform that allow people to understand a large scale of data over many years in seconds. To make full use of this feature, we visualize the popularity of 6 single years: 1950, 1960, 1970, 1980, 1990, and 2000 with Bookworm to understand the varying popularity of these years. Learned from Bookworm, publications mentioned each of these years most frequently in 1 to 3 years after them. For example, the popularity of 1950 peaked in 1952. Once the popularity reached its peak, it kept delinking in the following years.
Popularity of years: 1950, 1960, 1970, 1980, 1990, 2000 (Smoothing: 0 years）
For a point of verification, we also plot another group of years: 1950, 1951, 1952, 1953, 1954, and 1955. Here is the visual.
Popularity of years: 1950, 1951, 1952, 1953, 1954, 1955（Smoothing: 0 years）
From this visual, we can see more clearly what the popularity of a single year is like. The popularity of year A always rockets in 1-3 years after A itself. Then it decreases progressively. Decreasing rate is high at first and becomes lower later. Finally, the popularity maintains at a low level. Is this pattern universal? Does it reflect people’s memory and oblivion of history?
Other interesting findings with Bookworm
Each of them may shed light on several questions. For example:
What story does it tell for us? Is it attracting or persuasive?
What are the reasons behind? How come the features of the visual?
What are the visible conclusions and what are the invisible bias?
Streamgraph of “creativity” growth in the 20th century, by class
Word popularity of “Feminism” after 1750
Word Popularity of 4 countries after 1945
The search field
Currently, HT+BW supports one-word (unigram) entries as search terms, entered via the search field:
The facet fields display the options for faceted search. Initially, clicking "All texts" will pop up a menu that displays the possible facet fields. After a selection is made, you can modify it again by clicking on that label. Initially, on clicking "All" or inside this box, a popup menu displays a scrolling list of values that can be selected for the given facet. Multiple values for a given facet field can be chosen by clicking inside this box. Multiple values in a facet field constitute a logical OR for search purposes. So, for instance, in the example below, the search will be for "Genre:Biography OR Poetry". Different facet fields are in an implicit AND relationship with each other. In the figure below, three facet fields have been specified. The search criteria specified in this example translate into "(Class:unknown) AND (Genre: Biography OR Poetry) AND (Format: Book)."
Clicking on the Settings button, which looks like a gear, will display a menu for changing graph options to specify Time, Metric, Case and Smoothing.
Time range can be adjusted by dragging the "Time" slider for begin point and end point to the left and right as necessary.
This selection allows you to choose how the numerical values are counted. Depending on what option you choose, the label of the y-axis of the graph is changed accordingly, and the chart values adjusted. The four options available are as follows:
- % of words shows how frequently the unigram is used relative to all other tokens in the corpus, for the given year shown. This lets you see how often a word is used relative to the size of the corpus, without having to worry about things like whether there are more books in 1850 than 1900.
- % of texts gives the number of texts that use your search terms at least once as a proportion of the total number of texts published that year. Unlike "% of words," it will not be skewed by a single book that uses a word hundreds of times; however, it may be impacted by changing sizes.
- Word count plots the actual count of the searched word as the y-value for the plot.
- Text count plots a count as the y-value for the plot that is computed in the following way: only those volumes in which the searched word actually occurs are counted for creating the plot. So, each such volume registers as only a single count. (The word "text" is being used interchangeably with the word "volume".)
- Insensitive ignores the distinction between lowercase and uppercase characters when counting words
- Sensitive maintains the distinction between lowercase and uppercase.
Smoothing is a means to create a moving average over the data and to identify overall trends by removing jagged and discontinuous data points. Often trends become more apparent when data is viewed as a moving average. Smoothing windows are weighted: the year shown is weighted the most heavily, and the weights decrease in each direction until the smoothing span is reached. Smoothing options are described below:
- To see the raw data points, set smoothing to 0.
- To average one point on each side of a data point, set smoothing to 1, which counts the previous one, current one, and next one and divides that sum by 3.
- A smoothing setting of 5 means that 11 values will be averaged, 5 values on each side of the data point.
Accessing individual volumes in the HathiTrust Digital Library from HT+Bookworm
You can click on any point in the plot to see a listing of the volumes by decreasing order of contribution to the plot at that particular year. Each volume title is a hyperlink, clicking on which will take you to the corresponding volume in the HathiTrust Digital Library.
Michel, Jean-Baptiste, Shen, Y. K., Aiden, A. P., Veres, A., Gray, M. K., Pickett, J. P., ... & Aiden, E. L. "Quantitative analysis of culture using millions of digitized books." Science331.6014 (2011): pp. 176-182.
Unsworth, John. “Forms of Attention: Digital Humanities Beyond Representation," delivered at "The Face of Text: Computer-Assisted Text Analysis in the Humanities," The Third Conference of the Canadian Symposium on Text Analysis (CaSTA), McMaster University, November 19-21, 2004.
Gibbs, Frederick W., and Daniel J. Cohen. "A Conversation with data: prospecting Victorian words and ideas." Victorian Studies, Vol. 54, No. 1 (2011): pp. 69-77.