Tag Archives: KNIME

Looking for letters

PhD related, , , , , , , ,

In the posts I’ve written to date, I’ve learned some interesting things about my corpus of 40,000 news articles. I’ve seen how the articles are distributed over time and space. I’ve seen the locations they talk about, and how this shifts over time. And I’ve created a thematic index to see what it’s all about. But I’ve barely said anything about the articles themselves. I’ve written nothing, for example, about how they vary in their format, style, and purpose.

To some extent, such concerns are of secondary importance to me, since they are not very accessible to the methods I am employing, and (not coincidentally) are not central to the questions I will be investigating, which relate more to the thematic and conceptual aspects of the text. But even if these things are not the objects of my analysis, they are still important because they define what my corpus actually is. To ignore these things would be like surveying a large sample of people without recording what population or cohort those people represent. As with a survey, the conclusions I draw from my textual analysis will have no real-world validity unless I know what kinds of things in the real world my data represent.

In this post, I’m going to start paying attention to such things. But I’m not about to provide a comprehensive survey of the types of articles in my corpus. Instead I will focus on just one categorical distinction — that between in-house content generated by journalists and staff writers, and contributed or curated content in the form of readers’ letters and comments. Months ago, when I first started looking at the articles in my corpus, I realised that many of the articles are not news stories at all, but are collections of letters, text messages or Facebook posts submitted by readers. I wondered if perhaps this reader-submitted content should be kept separate from the in-house content, since it represents a different ‘voice’ to that of the newspapers themselves. Or then again, maybe reader’s views can be considered just as much a part of a newspaper’s voice as the rest of the content, since ultimately it is all vetted and curated by the newspaper’s editors.

As usual, the relevance of this distinction will depend on what questions I want to ask, and what theoretical frameworks I employ to answer them. But there is also a practical consideration — namely, can I even separate these types of content without sacrificing too much of my time or sanity? 40,000 documents is a large haystack in which to search for needles. Although there is some metadata in my corpus inherited from the Factiva search (source publication, author, etc.), none of it is very useful for distinguishing letters from other articles. To identify the letters, then, I was going to have to use information within the text itself. Continue reading Looking for letters →

What’s it all about? Indexing my corpus using LDA.

PhD related, , , ,

Months ago, I assembled a dataset containing around 40,000 Australian news articles discussing coal seam gas. My ultimate aim is to analyse these articles, along with other text data from the web, so as to learn something about the structure and dynamics of the public discourse about coal seam gas in Australia. I’m interested in dissecting how different parties talk about this topic, and how this ‘configuration’ of the public discourse changes over time.

Although I didn’t originally plan to, I’ve focussed much of my energy so far on exploring the geographic dimension of the news articles. I’ve looked at where the news has come from and what places it talks about. This is all important stuff to know when studying such a geographically defined issue as coal seam gas development. But I also need to know what is being talked about, not just where. Now, finally, I am ready to turn my attention to exploring the thematic content of the articles.

Well, almost. I’m ready, but the data isn’t. The dataset that I have been playing with all this time is stuffed with articles that I don’t want, and is missing many that I do. This is because the search parameters that I used to retrieve the articles from Factiva were very broad — I obtained every article that mentioned coal seam gas or CSG anywhere even just once — and because I applied a rather rudimentary method — keyword counts — for filtering out the less relevant articles. The dataset has served its purpose as a testing ground, but if I am to use it to actually say things about the world, I need to know what it contains. And more than that, I need the ability to customise what it contains to suit the specific questions that I decide to explore.

In other words, I need an index to my corpus. I need to know what every article is about, so I can include or exclude it at my discretion. In this post I’ll describe how I have created that index using a method of topic modelling called Latent Dirichlet Allocation, or LDA. Happily, this is the very same method that I was planning to use to analyse the thematic content of my corpus. So by creating an index for my corpus, I am already starting on the process of understanding what it’s all about. Continue reading What’s it all about? Indexing my corpus using LDA. →

Mapping concepts, comparing texts

PhD related, , , , , ,

In the previous post, I explored the use of function words — that is, words without semantic content, like it and the — as a way of fingerprinting documents and identifying sets that are composed largely of the same text. I was inspired to do this when I realised that the dataset that I was exploring — a collection of nearly 900 public submissions to an inquiry by the New South Wales parliament into coal seam gas — contained several sets of documents that were nearly identical. The function-word fingerprinting technique that I used was far from perfect, but it did assist in the process of fishing out these recycled submissions.

That exercise was really a diversion from the objective of analysing the semantic content of these submissions — or in other words, what they are actually talking about. Of course, at a broad level, what the submissions are talking about is obvious, since they are all responses to an inquiry into the environmental, health, economic and social impacts of coal seam gas activities. But each submission (or at least each unique one) is bound to address the terms of reference differently, focussing on particular topics and making different arguments for or against coal seam gas development. Without reading and making notes about every individual submission, I wanted to know the scope of topics that the submissions discuss. And further to that, I wanted to see how the coverage of topics varied across the submissions.

Why did I want to do this? I’ll admit that my primary motivation was not to learn about the submissions themselves, but to try my hand at some analytical techniques. Ultimately, I want to use computational methods like text analytics to answer real questions about the social world. But first I need some practice at actually doing some text analytics, and some exposure to the mechanics of how it works. That, more than anything else, was the purpose of the exercise documented below. Continue reading Mapping concepts, comparing texts →

The bottom-right cluster. All of these documents except Submission 0655 draw on the same template.

Using Junk words to find recycled text

PhD related, , , , , ,

Newton’s third law of motion — that for every action, there is an equal and opposite reaction — would appear to apply to the coal seam gas industry in Australia. The dramatic expansion of the industry in recent years has been matched by the community’s equally dramatic mobilisation against it. As my previous post showed, there are literally dozens of organisations on the web (and probably even more on Facebook) concerned in some way with the impacts of coal seam gas development. Some of these are well-established groups that have incorporated coal seam gas into their existing agendas, but many others seem to have popped up out of nowhere.

Most of these groups could be classified as community organisations insofar as they are concerned with a specific region or locality. But to think of them all as ‘grassroots’ organsiations, each having emerged organically on its own accord, might be a mistake. As the website network in my last post suggests, many of these groups might better be thought of as ‘rhizomatic’ (or lateral) offshoots inspired by the Lock the Gate Alliance. Lock the Gate emerged in 2010 and quickly reconfigured the landscape of community opposition to coal seam gas. Its campaigns, strategies and symbolism provided a handy template upon which locally focussed organisations could form. You’ll be hard-pressed to find a community-based anti-CSG group without a link to Lock the Gate on their website.

The lesson here is that voices that appear to be independent may to some extent be influenced or assisted by a small handful of highly motivated (or well resourced) groups or individuals. Having observed this possibility in the network of anti-CSG websites, I recently encountered it again while sifting through a very different dataset that I am preparing for  textual analysis. The dataset in question is the 893 public submissions that the Parliament of New South Wales received in response to its 2011 inquiry into the environmental, health, economic and social impacts of coal seam gas activities. The submissions came from all kinds of stakeholders, including community groups, gas companies, scientific and legal experts, government agencies, and individual citizens. Of particular interest to me were the 660 submissions from individual citizens. Here was a sizable repository of views expressed straight from the minds and hearts of individual people, undistorted by the effects of groupthink or coordinated campaigns. Or so I thought. Continue reading Using Junk words to find recycled text →