I. Required Programs TAACO Quick Start Guide: Windows 7 Kristopher Kyle and Scott Crossley 9-9- 2014 Before TAACO will work correctly, you must have Python 2.7, Natural Language Toolkit (NTLK), and NumPy installed, and you must download a few NLTK data libraries. If you regularly use Python and NLTK and have successfully installed NumPy, ensure that you have downloaded the NLTK data libraries maxent_treebank_pos_tagger, punkt, wordnet, and wordnet_ic, then continue to the How to Use TAACO section. If you are new to Python, or have not successfully installed NLTK and/or NumPy, please follow the directions below for your operating system. Windows (Tested on Windows 7 64 bit) 1. Download and install Enthought Canopy Express: https://store.enthought.com/ (see Figures 1 and 2). Your processor may be 32 bit or 64 bit. Canopy Express is a distribution of Python 2.7 that comes with a number of important Python packages pre- installed. These packages can be installed individually, but this tends to take quite a bit of time and computer expertise. Installing Canopy Express saves you quite a bit of time and effort. Figure 1. Enthought Canopy download screenshot
Figure 2. Click Run to install the program 2. After you have downloaded and installed Canopy Express, open the program and click through the default choices, including making Canopy your default Python environment. 3. After setting up Canopy Express, close it. 4. To download and install NLTK, open the program Command Prompt (you may have to use the Windows Search function under All Programs to find this program), type (do not copy and paste!) pip install U nltk, and then press return (see Figure 3).
Figure 3. Typing the command into Command Prompt to install NLTK 5. Download the NLTK data libraries. a. Type python in Command Prompt and then press return (see Figure 4). Figure 4. Typing python in Command Prompt to open Python
b. Then type import nltk and press return (see Figure 5). Figure 5. Typing import nltk and nltk.download() to import nltk and then open the download prompt c. Then type nltk.download() and press return, which will open a program that assists you in downloading the data libraries (see Figure 5). d. The easiest course is to go to download all packages (if you click the download button with All packages highlighted, this will download all packages; see Figure 6). If you are worried about hard drive space, you can choose to download only the following packages (see Figure 7): - maxent_treebank_pos_tagger : Treebank POS Tagger (maximum entropy) - punkt : Punkt Tokenizer Models - wordnet : WordNet - wordnet_ic : WordNet- InfoContent
Figure 6. Main screen of the NLTK Downloader Figure 7. Selecting specific packages to install
6. Proceed to the How to Use TAACO steps below How to Use TAACO: Once you have installed Python and NLTK, TAACO should function correctly. To use TAACO, follow these steps: 1. Download TAACO from http://www.kristopherkyle.com/taaco.html 2. Unzip the TAACO application (if it is still compressed) 3. Double- click the TAACO icon 4. You may get a warning regarding the TAACO application not being safe. This is only because we have not gone through the process of registering TAACO with Microsoft. We assure you our tool is perfectly safe. 5. If you cannot see a box that says Waiting for Data to Process, expand the size of the program until you can see this box. 6. Select the folder where the files you wish to process are (these files must be in.txt format) by clicking the Select Input Folder button. 7. Select the folder where you wish to save the output file by clicking the Select Output Folder button. 8. It is advised to type a unique filename into the output file box (to prevent a file being overwritten) and then click the Submit Filename button. 9. To start the program, click the Process Texts button. TAACO s progress will be displayed at the bottom of the program. 10. To interpret your results, reference the accompanying TAACO Index Spreadsheet.