Tools

Visualization

Paper Machines

Paper Machines is a topic modelling and visualization tool available as a plugin for Zotero. It analyzes Zotero bibliographic collections based on a selection of text mining processes, and enables users to export a variety of visualizations, such as word clouds, phrase nets or heat maps, from the results.

Rated 0.0 out of 5.

Search

Overview

Overview is a free, web-based tool for document mining. Developed by and for journalists, Overview is also broadly applicable to other corpus-based research applications. Aside from providing advanced search features such as regular expressions and fuzzy matching, Overview automatically sorts documents according to topic and provides a fast visualization and reading interface. You can apply tags to search results, individual documents, or entire topics to track what you've found, then export the entire document set with its tags to perform further processing or create visualization. It’s been used for qualitative research, social media conversation analysis, legal document review, digital humanities, and more. Overview can be run via the web interface (max 200 000 documents) or downloaded and installed. The system can import documents in PDF or CSV format and currently supports analysis in 9 languages. The code is open source and there is a plugin API to extend Overview's analysis and visualization capability.

Rated 0.0 out of 5.

Visualization

RAW

RAW is a free, open-source web application for creating visualizations that can be exported in SVG or PNG format and embedded into a website. Users can upload data with RAW's web interface and generate a variety of vector-based visualizations. RAW uses the D3.js library, and is maintained by the Density Design research studio at Polaitecnico di Milano.

Rated 0.0 out of 5.

Analysis

SentiStrength

SentiStrength is a tool for sentiment analysis available for free to academic users (with registration), in Java and Windows-optimized versions. It estimates the strength of sentiment in short texts, and can handle informal language. Strength is expressed in a scale of 1 to 5 for positive sentiments, and -1 to -5 for negative; it also includes an option for assessing neutrality. SentiStrength is available for multiple languages, with extensibility for languages and contexts not already included. According to the developers, SentiStrength has human-level accuracy for short English texts. Commercial users must purchase a license.

Rated 0.0 out of 5.

Analysis

Compare Lists

Compare Lists is a free, web-based tool for comparing text from two lists of URLs supplied by the user. This tool was developed and is maintained by the Digital Methods Initiative.

Rated 0.0 out of 5.

Search

Google Blogsearch Scraper

The Google Blogsearch Scraper is a free, web-based tool for batch querying Google Blog Search. Users can submit a list of URLs to search, and a set of keywords to apply to them. For each query, the tool supplies a tag cloud and an HTML table, plus a text file that may be downloaded for further analysis.

Rated 0.0 out of 5.

Search

Netvizz

Netvizz is a free, web-based tool for extracting datasets from Facebook. As the tool utilizes Facebook Apps, users must have a Facebook account to access it.

Rated 0.0 out of 5.

Analysis

NetvizzToSentiStrength

NetvizzToSentiStrength is a free, web-based sentiment analysis tool for short texts, designed to work particularly with a Netvizz search of Facebook data. In addition to Netvizz data, it also accepts DMI-TCAT and CSV files.

Rated 0.0 out of 5.

Search

Google News Scraper

Google News Scraper is a free, web-based tool for batch querying news.google.com. Users can specify search terms, news sources, date range, location, language, version of Google (ex. .com, .ca, .co.uk, etc.), and what facets to include in the output. The tool returns the results as a table suitable for further analysis, and can also include thumbnail images for each search item.

Rated 0.0 out of 5.

Search

Googlescraper (Lippmannian Device)

Googlescraper (Lippmannian Device) is a free, web-based tool for querying Google and saving the results for analysis. It allow users to specify URLs, keywords, and which keywords apply to each URL. The tool outputs as HTML, a tag cloud, and a text file. The tool must be used with FireFox, and requires users to first install the DMI toolbar , which permits the tool to work around Google's block against automated requests.

Rated 0.0 out of 5.