The Phone Record Analyzer is a tool for the automatic analysis of phone record data. Phone records typically come in a standard format and the information we're often looking to garner from them is certainly standardized. What do we want? We want to know who our targets are calling, how often they're talking, when they're talking, and which numbers are common amongst all of our targets if we're looking at more than one person.

This tool accomplishes all of that automatically for you so that you don't have to waste time writing the same function calls in Excel every time you're reviewing phone records. Once your data is loaded, the app will generate at least two analyses: 1) a Call Frequency report, which is a data table that holds the frequencies of outgoing and incoming calls for every number in the phone records, and 2) a graphical breakdown of the calls by day for any given month in any given year. It will produce an additional analytical product if you have phone records for more than one person, which is a link analysis that shows all of the telephone numbers that your targets have in common. In other words, let's say you have two people for whom you collected phone records: John Doe and Jane Smith. One of the things we want to know is whether or not John and Jane are talking to the same people. The app will determine that for you and then show you that information in a link analysis chart.

All analytical products, including data tables, charts, and graphs, can be exported for further use.

First and foremost, you need to have R and RStudio installed. If you do not have these things installed, please Google how to do this.

There are two data formats that this tool can accept: TXT and CSV. If your data are currently in PDF or an Excel-specific format, you will want to export the data to one of the accepted formats. For PDFs, this will mean opening the file Acrobat Reader and choosing the "Save as Other" option and then navigating to the "Text (Plain)" option. For Excel, this means combining all of your data into one sheet and saving it as a "Comma-Separated Values (CSV)" format. When loading the data, the app will ask two questions that are used for determining how to process the data. The first question asks what type of compliance you're dealing with, and the second asks how your file was created. This second question is important because the TXT file that is created when saving a PDF is different from the TXT file that is provided from the phone company. Make sure you answer this question accurately. If you're uploading a CSV, this question doesn't matter so just choose the "I made it myself" option.

There is one more thing worth noting about CSV files. Because the app is not parsing the data itself, it needs the data to be in a very specific format in order to work properly. This means that if you're uploading a "Formatted CSV", your data must contain certain columns and they must be labeled properly. These columns include: Target, Number_Dialed, and Date. The "Target" column is a the number of the person for whom you ordered the phone records. The "Number_Dialed" column contains the numbers that your target(s) have spoken to and "Date" is the date the call took place. This is the bare minimum amount of information that must be included. Another piece of information that you can include that enhances the analyses is "Direction", which is a column indicating if the call was outgoing or incoming. If you're uploading a "Raw CSV" file, the formatting is a little looser as the app will try to identify and format the columns it needs. However, this is patchy at best and a formatted CSV is preferred at this time.

Currently, if you're looking to take advantage of the app's parsing functionality, it can only parse TXT documents that are generated from AT&T compliance. Functionality for other phone companies will be coming in the future, but as for now it is only for AT&T. If you're uploading a CSV, it doesn't matter which company the data came from.

Any issues that you encounter or feedback that you have should be filed using the "Issues" link found on the right-hand side of the screen.