Assignment 4: Entity Extraction

взять займ безработному. For this assignment you will evaluate the performance of OpenCalais, a commercial entity extraction service, against your hand annotations.

1. Pick five random news stories and hand-annotate them. Pick an English-language news site with many stories on the home page, or a section of such a site (business, sports, etc.) Then generate five random numbers from 1 to the number of stories on the page. Cut and paste the text of each article into a separate  file, and save as plain text (no HTML, no formatting.)

2. Detect entities by hand in each article. Paste the text of each article into an RTF or Word document and go through it, underlining every entity. Count every mention of a person, place, or organization, including alternate names (“the president”) and pronoun references. Count how many entity references appear in each document (multiple mentions of the same entity all count.)

3. Now run each document through OpenCalais. You can paste the text into the demo page here. Now count:

  • Your hand count of entity references
  • Correct entities: entities found by OC and marked by you
  • Incorrect entities: entities found by OC but not marked by you
  • Missed entities: entities marked by you but not found by OC

4. Turn in:

  • Your hand-marked documents.
  • A spreadsheet. Please turn in a spread sheet with one row per document, and four columns: hand-labelled, correct, incorrect, missed
  • Make sure your spreadsheet includes totals of these four columns across all documents
  • Your analysis. Report on any patterns in the failures that your see. Where is OpenCalais most accurate? Where is it least accurate? Are there predictable patterns to the errors? Are there ambiguities as to what is really an entity?
This assignment is due before class on Friday,  December 1.

Assignment 3: Analyzing Stop and Frisk for Racial Bias

For this exercise, you will break into three groups. Two will analyze data, and the third will do legal research. The data for this assignment is available here.

All groups: How can we encode our ideas about racial fairness into a quantitative metric? That is the fundamental question underlying this assignment and you must answer it. Don’t build models that give you answers to useless questions; each model you build must embody some justifiable concept of fairness. Many such metrics have been proposed; part of the assignment is researching and evaluating them. I’ve proposed some models that might be interesting, but I’ll be just has happy — perhaps happier — to have you tell me why these particular model formulations will not yield an interesting result. Also, you have to tell me what your modeling results mean. Is there bias? In what way, how significant is it, and what are alternate explanations? Uninterpreted results will not get a passing grade.

Group 1: Analyze the stop and frisk data using a multi-level linear model

This group will analyze the data along the lines of Gelman 2006. However, that paper used Bayesian estimation whereas this team will used standard linear regression. Use RStudio with the lme4 package, as described in this tutorial.

As above, you need to choose and justify quantitative definitions of fairness. Should you look at stop rate? Hit rate? And should you compare to the racial composition of each precinct? Or per-race crime rates? And if so, which crimes?

Each conceptual definition of fairness can be embodied in many different statistical models. Compare at least two different statistical formulations. For example you might end up modeling:

  • hit rate vs race, per precinct
  • hit rate vs race, per precinct, with precinct random effects

You must also choose the unit of analysis. Is precinct or census block a better unit? Why? Or you could compare the same model on different units.

Your final post must include a justification of the metric and model choices, and useful visualization of the results, and interpretation of both the regression coefficients and uncertainty measures (standard errors on regression coefficients, and modeling residuals.)

Group 2: Analyze the stop and frisk data using a Bayesian model

This group will analyze the data along the lines of Simoiu 2016, by adapting their code. For a tutorial on how to set up Bayesian modeling in R, see Bayesian Linear Mixed Models using STAN.

As with group 2, you must research and choose a definition of fairness, fit at least two different statistical formulations, and interpret your results including the uncertainty (posterior distributions and residuals.) Be sure to visualize your model fit, as in figure 7 of Simoiu.

I would be happy to see you replicate the threshold test of Simoiu. However, I want you to explain why the threshold test makes sense as a fairness metric. If it doesn’t make sense I want you to design a new model. Perhaps the assumption that each officer can make correctly calibrated estimates of the probability that someone is carrying contraband is unrealistic, and your model should be based on the idea that the estimates are biased and try to model that bias as latent variables.

Group 3: Research legal and policy precedent for statistical tests

This group will scour the legal literature to determine what sorts of statistical tests have been used, or could be used, to answer legal questions of discrimination. You will also research the related policy literature: what sorts of tests have governments, companies, schools etc. used to evaluate the presence or significance of discrimination. For an entry into the literature, you could do worse than to start with the works referenced by Big Data’s Disparate Impact.

This group will not be coding, but I expect you to ask not only what fairness metrics might be appropriate (as the other groups must also do) but 1) whether or not these metrics might hold up in court and 2) whether they have ever been used outside of court.

And one particular question I would like you to answer: would Simiou’s “threshold test” have legal or policy relevance?


Due Friday Nov 10

Assignment 2: Filter Design

For this assignment you will design a hybrid filtering algorithm. You will not implement it, but you will explain your design criteria and provide a filtering algorithm in sufficient technical detail to convince me that it might actually work — including psuedocode.

1. Decide who your users are. Journalists? Professionals? General consumers? Someone else?

2. Decide what you will filter. You can choose:

  • Social network updates, like the Facebook news feed or Twitter trending topics
  • A news organization recommendation engine
  • The whole web, like Google News
  • something else

3. List all available information that you have available as input to your algorithm. If you want to filter Facebook or Twitter, you may pretend that you are the company running the service, and have access to all posts and user data — from every user. You also also assume you have a web crawler or a firehose of every RSS feed or whatever you like, but you must be specific and realistic about what data you are operating with.

4. Argue for the design factors that you would like to influence the filtering, in terms of what is desirable to the user, what is desirable to the publisher (e.g. Facebook or a news organization), and what is desirable socially. Explain as concretely as possible how each of these (probably conflicting) goals might be achieved through in software. Since this is a hybrid filter, you can also design social software that asks the user for certain types of information (e.g. likes, votes, ratings) or encourages users to act in certain ways (e.g. following) that generate data for you.

5. Write psuedo-code for a function that produces a “top stories” list. This function will be called whenever the user loads your page or opens your app, so it must be fast and frequently updated. You can assume that there are background processes operating on your servers if you like. Your psuedo-code does not have to be executable, but it must be specific and unambiguous, such that I could actually go and implement it. You can assume that you have libraries for classic text analysis and machine learning algorithms. So, you don’t have to spell out algorithms like TF-IDF or item-based collaborative filtering, or anything else you can dig up in the research literature, but simply say how you’re going to use such building blocks. If you use an algorithm we haven’t discussed in class, be sure to provide a reference to it.

6. Write up steps 1-5. The result should be no more than three pages. You must be specific and plausible. You must be clear about what you are trying to accomplish, what your algorithm is, and why you believe your algorithm meets your design goals (though of course it’s impossible to know for sure without testing; but I want something that looks good enough to be worth trying.)

Due before class, October 13

Assignment 1: Topic Modeling

займ без процентов на карту. This assignment is designed to help you develop a feel for the way topic modeling works, the connection to the human meanings of documents, and common ways of handling a time dimension. You will analyze the State of the Union speeches corpus, and report on how the subjects have shifted over time in relation to historical events.

1. Download the source data file state-of-the-union.csv. This is a standard CSV file with one speech per row. There are two columns: the year of the speech, and the text of the speech. You will write a Python program that reads this file and turns it into TF-IDF document vectors, then prints out some information. Here is how to read a CSV in Python. You may need to add the line


to the top of your program to be able to read this large file.

The file is a csv with columns year, text. Note: there are some years where there was more than one speech! Design your data structures accordingly.

2) Feed the data into gensim. Now you need to load the documents into Python and feed them into the gensim package to generate tf-idf weighted document vectors. Check out the gensim example code here. You will need to go through the file twice: once to generate the dictionary (the code snippet starting with “collect statistics about all tokens”) and then again to convert each document to what gensim calls the bag-of-words representation, which is un-normalized term frequency (the code snippet starting with “class MyCorpus(object)”

Note that there is implicitly another step here, which is to tokenize the document text into individual word features — not as straightforward in practice as it seems at first, but the example code just does the simplest, stupidest thing, which is to lowercase the string and split on spaces. You may want to use a better stopword list, such as this one.

Once you have your Corpus object, tell gensim to generate tf-idf scores for you like so.

3) Do LSI topic modeling. You can apply LSI to the tf-idf vectors, like so. You will have to supply the number of topics to generate. Figuring out a good number is part of the assignment. Print out the resulting topics, each topic as a lists of word coefficients. Now, sample ten topics randomly from your set for closer analysis. Try to annotate each of these ten topics with a short descriptive name or phrase that captures what it is “about.” You will likely have to refer to the original documents that contain high proportions of that topic, and you will likely find that some topics have no clear concept.

Turn in: your annotated topics plus a comment on how well you feel each “topic” captured a real human concept.

4) Now do LDA topic modeling. Repeat the exercise of step 3 but with LDA instead, again trying to annotate ten randomly sampled topics. What is different?

Turn in: your annotated topics plus a comment on how LDA differed from LSI.

5) Now choose one of the following exercises:

a) Come up with a method to figure out how topics of speeches have changed over time. The goal is to summarize changes in the State of the Union speech in each decade of the 20th and 21st century. There are many different ways to use topic modeling to do this. Possibilities include: visualizations, grouping speeches by decade after topic modeling, and grouping speeches by decade before topic modeling. You can base your algorithm on either LSI or LDA, whichever you feel gives the most insight. Choose a method, then use your decade summarization algorithm to understand what the content of speeches was in each decade.

Turn in: a description of your decade summarization algorithm, and an analysis of how the topics of the State of the Union have changed over the decades of the 20th century. What patterns do you see? Can you connect the terms to major historical events? (wars, the great depression, assassinations, the civil rights movement, Watergate…)

or b) Analyze a different document set. Try LDA on a different document set, this collection of AP wire stories. Repeat the process of choosing the number of topics, fitting a model, and interpreting a random sample of 10 of them. Are the topics any clearer on this document set? If so, why? You may wish to look at previous LDA results on these documents, the top 20 words from 100 topics.

Turn in: your annotated topic sample, plus a description of the differences between the output on these documents vs. the State of the Union documents. Does one work better than the other? If so, define “better.”

This assignment is due Friday, September 29 at 10:00 AM. You may email me the results.

Syllabus Fall 2017

The course is a hands-on, research-level introduction to the areas of computer science that have a direct relevance to journalism, and the broader project of producing an informed and engaged public. We study two big ideas: the application of computation to produce journalism (such as data science for investigative reporting), and journalism about areas that involve computation (such as the analysis of credit scoring algorithms.)

Alon the way we will touch on many topics: information recommendation systems but also filter bubbles, principles of statistical analysis but also the human processes which generate data, network analysis and its role in investigative journalism, visualization techniques and the cognitive effects involved in viewing a visualization.

Assignments will require programming in Python, but the emphasis will be on clearly articulating the connection between the algorithmic and the editorial.

Research-level computer science material will be discussed in class, but the emphasis will be on understanding the capabilities and limitations of this technology. Students with a CS background will have opportunity for algorithmic exploration and innovation, however the primary goal of the course is thoughtful, application-oriented research and design.

Format of the class, grading and assignments.
This is a fourteen week, six point course for CS & journalism dual degree students. (It is a three point course for cross-listed students, who also do not have to complete the final project.) The class is conducted in a seminar format. Assigned readings and computational techniques will form the basis of class discussion. The course will be graded as follows:

  • Assignments: 40%. There will be five homework assignments.
  • Final project 40%: Dual students will be complete a medium-ish final project (others will have this 40% from assignments)
  • Class participation: 20%

Assignments will involve experimentation with fundamental computational techniques. Some assignments will require intermediate level coding in Python, but the emphasis will be on thoughtful and critical analysis. As this is a journalism course, you will be expected to write clearly. The final project can be either a piece of software (especially a plugin or extension to an existing tool), a data-driven story, or a research paper on a relevant technique.

Dual degree students will also have a final project. This will be either a research paper, a computationally-driven story, or a software project. The class is conducted on pass/fail basis for journalism students, in line with the journalism school’s grading system. Students from other departments will receive a letter grade.

Week 1: Introduction and Clustering – 9/8
First we ask: where do computer science and journalism intersect? CS techniques can help journalism in two main ways: using computation to do journalism, and doing journalism about computation. We’ll spend most of our time on the former: data-driven reporting, story presentation, information filtering, and effect tracking. Then we jump right into clustering and the document vector space model, which we’ll need to study filtering.


Viewed in class

Week 2: Filtering Algorithms – 9/15
The filtering algorithms we will discuss this week are used in just about everything: search engines, document set analysis, figuring out when two different articles are about the same story, finding trending topics. The main topics are matrix factorization, probabilistic topic modeling (ala LDA) and more general plate-notation graphical models, and word embeddings. Bringing it to practice we will look at Columbia Newsblaster (a precursor to Google News) and the New York Times recommendation engine.



Discussed in class

Assignment:  LDA analysis of State of the Union speeches.

Week 3: Filters as Editors – 9/22
We’ve studied filtering algorithms, but how are they used in practice — and how should they be? We will study the details of several algorithmic filtering approaches used by social networks, and effects such as polarization and filter bubbles.



Viewed in class

Week 4: Computational Journalism Platforms   – 9/29
We introduce the Overview document mining system and the Computational Journalism Workbench. Then we develop pitches for final projects, which may include writing plugins for these systems.

Guest Speaker: Alex Spangher, New York Times



Assignment – Design a filtering algorithm for an information source of your choosing

Week 5: Quantification, Counting, and Statistics – 10/6
Every journalist needs a basic grasp of statistics. Not t-tests and all of that, but more grounded and more practical. How do we know we’re measuring the right thing? Why are we doing stats at all? Then a journalism oriented tutorial on the fundamental ideas of probability, conditional probability, and Bayes’ theorem.



No class 10/13

Week 6: Inference – 10/20
There is a long history of fields grappling with the problem of determining truth in the face of uncertainty, from statistics to intelligence analysis. We’ll start with statistics, the notion of randomness that is so crucial to the idea of statistical significance. Then we’ll talk about determining causality, p-hacking and reproducibility, and the more qualitative, closer-to-real-world method of analysis of competing hypothesis.



Viewed in class

Week 7: Discrimination and Algorithmic Accountability – 10/27
Two topics this week. Discrimination is an important topic for reporters and for society, but analyzing discrimination data is more subtle and complex than it might seem. Algorithmic accountability is the study of the algorithms that regulate society, from high frequency trading to predictive policing. We’re at their mercy, unless we learn how to investigate them.



Assignment: Analyze NYPD stop and frisk data for racial discrimination.

No class 11/3

Week 8: Visualization, Network Analysis – 11/3
Visualization helps people interpret information. We’ll look at design principles from user experience considerations, graphic design, and the study of the human visual system. Network analysis (aka social network analysis, link analysis) is a promising and popular technique for uncovering relationships between diverse individuals and organizations. It is widely used in intelligence and law enforcement, and inreasingly in journalism.


  • Visualization, Tamara Munzner
  • Network Analysis in Journalism: Practices and Possibilities, Stray



Assignment: Compare different centrality metrics in Gephi.

Week 9 Knowledge representation – 11/10
How can journalism benefit from encoding knowledge in some formal system? Is journalism in the media business or the data business? And could we use knowledge bases and inferential engines to do journalism better? This gets us deep into the issue of how knowledge is represented in a computer. We’ll look at traditional databases vs. linked data and graph databases, entity and relation detection from unstructured text, and traditional both probabilistic and propositional formalisms. Plus: NLP in investigative journalism, automated fact checking, and more.



Viewed in class

Assignment: Text enrichment experiments using OpenCalais entity extraction.

Week 10: Truth and Trust – 11/17
We went through The Ethics of Persuasion slides.
Computational propaganda. Structure of information operations. Fake news detection and tagging. Credibility schema. Systems to detect and combat abuse and harassment.

Speaker: Ed Bice, Meedan



No class 11/24

Week 11: Privacy, Security, and Censorship – 12/1
Who is watching our online activities? Who gets to access to all of this mass intelligence, and what does the ability to survey everything all the time mean both practically and ethically for journalism? In this lecture we cover both the basics of digital security, and methods to deal with specific journalistic situations — anonymous sources, handling leaks, border crossings, and so on.


  • Digital Security for Journalists, Part 1 and Part 2, Stray


Week 12: Tracking flow and impact – 12/8
How does information flow in the online ecosystem? What happens to a story after it’s published? How do items spread through social networks? We’re just beginning to be able to track ideas as they move through the network, but it’s still very difficult to really measure the public interest impact of journalism.



Week 13: Final Project Presentations – 12/15