How to use WordValue

1. Create an account

First-time users need to create an account to access WordValue. This only requires a self-selected username, a self-selected password and the indication of an e-mail address. To delete your account, simply press the DELETE button in the account settings.

2. Combine a search table and a target text

To use WordValue, you need two files: a table containing a list of search words (and usually some additional information) and a text file in which the words from the table will be searched.

You can try out WordValue right away with several preset text files (e.g. The happy prince by Oscar Wilde) and several preset tables (e.g. the 200 most frequent English words) by going straight to the RESULTS tab.

Otherwise you can upload your own files:

2.1 Upload a text file that you wish to analyse

Go to the TEXTS tab and click the button NEW TEXT. Then select and upload a text file from your computer. The existing filename is used by default.

Please note that uploaded text files need to be in .txt format with UTF-8 encoding. You can easily convert a Word document into a .txt file by saving a copy of it and changing the file type option to .txt. Then click Tools in the same window. Select "Web options", go to the tab Encoding and select "Unicode (UTF-8)" in the dropdown menu "Save this document as".

You also have the option to enter text directly into a text box and save it with a filename. In that case the text is automatically saved in UTF-8 format.

2.2 Upload a table with word-related information that is to be mapped onto the words in the text

Go to the tab TABLES and click the button NEW TABLE. Then select and upload a new table from your computer. The table can be either an Excel spreadsheet or a .csv file. Please note that uploaded .csv files should use UTF-8 encoding.

A possible table may look like this:

keys Part_of_speech Rank Frequency Length_letters
the determiner 1 6187267 3
be verb 2 4239632 2
of preposition 3 3093444 2
and conjunction 4 2687863 3

The first line of each column in the spreadsheet needs to specify the attribute (i.e. the category of information contained within the column), e.g. rank or frequency of the word. The first line thus acts as a header. The example table shown here avoids spaces in these labels (as is customary), but WordValue also accepts spaces.

The first column in the spreadsheet needs to contain the keys (i.e. the word forms that are searched in the text, and onto which the information from the other columns is mapped), e.g. the or be. The other columns specify the values that are linked to the individual keys, e.g. the part of speech determiner or the rank 1.

The first cell in the spreadsheet needs to be called keys. Note that each key needs to be unique and can only be assigned a single code for each attribute. Search keys may contain internal spaces, hyphens and even line breaks. Keys in Excel spreadsheets may also contain commas (in contrast to keys in .csv files, where this is not possible).

Please note that even if you only want to use WordValue to search for a list of words in a target text, the software requires a search table. This means that you need a second column with a dummy value of your choice (e.g. the value 1 in all cells).

While there are no restrictions on the size of the texts and tables, the time required for the first run of the experiment increases with both documents’ number of lines, as WordValue automatically computes the frequency of all keys from the table and of all the tokens in the text for further use. Please make sure to use Excel files with a single worksheet and a finite number of columns with content.

If you want to see the content of a text or table while using WordValue, go to the tabs TEXTS or TABLES and click on the filename to see the content of the corresponding text or table. To delete files, use the DELETE button next to the filename.

3. Customize your search options to obtain the best results

In the tab RESULTS, select a text and a table from the drop-down menus and press LOAD RESULTS. Then adapt the search options to your requirements.

You can either search your texts for the keys exactly as they are in your spreadsheet, or you can ignore case (and thus also find capitalised sentence-initial keys), or you can select the lemmatised search, which also retrieves inflected forms of your keys (e.g. the key game also matches games in your text). Lemmatised searches are always case-insensitive.

4. Select the required output type

The results of your search can be viewed either as a table with frequency information or in the innovative rainbow text format. Check the corresponding box to view the relevant output type. If both output types are selected, the frequency table is shown on top.

The rainbow text format permits a quick and intuitive overview of the values of relevant words in the text itself. The attributes are mapped directly onto each word in its context by using colour-coding. If this option is selected, further choices are available.

5. Select colours

In the first step, you need to select the attribute (i.e. the column from your spreadsheet) for which you wish to carry out the colour-coding. In the second step, you need to specify whether the attribute is categorical, ordinal or numerical.

For categorical attributes like PART OF SPEECH (whose values NOUN, ADJECTIVE, VERB etc. have no inherent ordering), WordValue automatically generates a list of all the value types in the corresponding column and asks you to select a colour for each of these values by interacting with a colour picker offering a large number of colours, whose precise form depends on the browser used.

Ordinal attributes like COMPARISON (whose values POSITIVE, COMPARATIVE and SUPERLATIVE are ordered but not equally spaced) can be colour-coded like categorical variables with the additional use of an intuitive systematic colour-coding scheme going from warm to cold colours, from light to dark colours etc.

For numerical attributes like word frequency, WordValue offers the option to map the values to a colour gradient, for which users can define the start and end colour. WordValue then automatically retrieves all unique values of that attribute and maps these values onto the colour gradient.

Please note that the attribute selected by default is the numerical KEY FREQUENCY (i.e. how often the search keys occur in the target text). This means that if you select an ordinal or categorical attribute while still in numerical mode, a warning message will pop up to remind you that the attribute you selected can not be interpreted as numerical. In that case, please choose ordinal or categorical mode, whichever you prefer for your use-case. Alternatively, you can also select ordinal or categorical mode before you change the selected attribute, since any numerical attribute can also be interpreted as ordinal or categorical.

In addition, it is possible to modify the colour of all words in the text that do not match any search key. Changing this "background colour" to white permits a unique focus on the targeted items, whereas e.g. a light grey provides unobtrusive contextual information.

6. View and export the results

For convenient viewing, the results table and the rainbow-coloured text adapt dynamically to any changes in the parameters.

The results table can be easily sorted by clicking on the header of the corresponding column. Click a second time to reverse the ordering of the search keys in the first column, the search key frequencies in the second column and the values in any other column. The table with the results can be downloaded in .csv and Excel format by clicking on the corresponding buttons.

The rainbow-coloured texts are displayed in html format. This permits easy copy-pasting into Microsoft Word (and other appropriate software), where they can be saved and further processed.