Posts Tagged ‘sentimentanalysis’

Stock Picker Shows the Potential of Leading Indicator Pattern

A recent blog over at MIT’s Technology Review site caught my attention. Who wouldn’t read a post when the title promises “AI That Picks Stocks Better than the Pros”? I was expecting to learn about cutting-edge research into neural networks or some such, but instead I found a description of an approach I have been pitching for years now.

The “Arizonal Financial Text System” (AZFinText) works by “ingesting large quantities of financial news stories (in initial tests, from Yahoo Finance) along with minute-by-minute stock price data, and then using the former to figure out how to predict the latter”. The stories get a unique type of sentiment-analysis treatment, which was written by AZFinText’s authors, Robert P. Schumaker of Iona College in New Rochelle and and Hsinchun Chen of the University of Arizona. (more information can be found in their IEEE paper)

As you might expect, “Bad” stories can be expected to make a stock go down, while “Good” stories can make it go up. But what differs with Schumaker and Chen’s approach is that they did not use traditional human gauges of sentiment. Rather than look for emotionally weighted words like hate, love, good, or bad used in a typical sentiment analysis technique (yes, I am oversimplifying the process greatly), they back-tested the moves in a stock’s price against historical stories and used that data to derive the words that seemed to influence movement. This uncovered terms like hereto, comparable, charge, summit and green which caused the stock to move down, while words like planted, announcing, front, smaller and crude triggered an increase.

Before I explain my similar experiments in this area using mashups, I should point out that the basic idea here is nothing new. The article sites research in this area going back at least to 1990.

In my book, I describe applying mashups to this problem in my explanation of the Leading Indicator pattern:

Changes in a leading indicator may foretell downstream impact in other areas. Some, like the measurement of new home sales or bankruptcy filings have an obvious affect on many businesses. Well- known leading indicators do not offer any particular competitive advantage. Finding previously unknown relationships that predict business cycles and trends can be extremely useful.

Schumaker and Chen used Yahoo News stories in their study. But of course, the more data you use, the more onerous the collection problem becomes. As I noted in Mashup Patterns:

The key to uncovering these hidden links begins with collecting Time Series data from multiple locations. But with each additional source that is surveyed, the complexity of the data gathering process increases.

I also describe how collecting this data is applicable for discovering both Leading and Coincident Indicators. For example:

…knowledge of the week’s upcoming television schedule might not seem useful for a chain of pet stores. But when matched against a set of key words, a time series of “pet-themed” broadcasts can be assembled. Mashed up against a database of customer purchases, a Leading Indicator might emerge between Dog Shows and increased sales. The retailer now has a mechanism for advance inventory and advertising planning.

But before this post “goes to the dogs”, let’s get back to the Technology Review article. The reason it really stuck with me is because I had already demonstrated working implementations. About a year ago, I was invited to speak with several hedge funds, and for two of them I demonstrated a mashup that used the New York Times news API along with Reuter’s Open Calais to forecast stock fluctuations.

Although it is not noted how Schumaker and Chen got their data from Yahoo, using the data collection capabilities of many mashup products (for example, Kapow, Convertigo, or Connotate) are an excellent source for your own Leading Indicator implementation.

The key challenge is to stop thinking about external business sites from a consumer’s perspective and instead view them as databases to be mined via periodic data extraction. Here’s a final example from the Leading Indicator chapter that I know at least two firms have actually implemented:

Consider the value of building a mashup against popular online travel sites. Each day, an automated agent could book multiple flights between New York and London and from Boston to San Francisco. The mashup emulates the customer booking experience all the way up through seat selection, at which point it records the number of seats available and the ticket price. Naturally, the mashup wouldn’t complete the process of paying for the trip.

This information can be used to extrapolate the performance of the sector in advance of quarterly reports. Lots of seats available even as ticket prices decline? Probably not a sign of good earnings. Regularly filled seats at soaring prices might seem like good news until a time series of fuel prices hitting historical highs is added.

AZFinText operates on an extremely tight window of time, according to the article. It attempts to find stories that will move a stock within the next 20 minutes or so. This means that to be of practical use, it probably needs to be part of an automated trading system. But I hope the other examples I’ve shown demonstrate that larger timeframes can be involved and there can be plenty of time for actual people to evaluate the results and decide on a course of action. The Leading Indicator pattern might help you get information that your competitors don’t have. Or if they will have it eventually, you can at least get this knowledge before they do.

  • Zillow Home Valuation GetSearchResults
    The GetSearchResults API allows developers to find Zillow listings for a property for a specified address. The aim of the returned content is to contain the address for the property or properties, the Zillow Property ID (ZPID), current Zestimate®, the date the Zestimate was computed, a valuation range and the Zestimate ranking for the property […]
  • Zillow Neighborhood GetRegionChildren
    The GetRegionChildren API aims to return a list of subregions with the following information for a specified region: Subregion Type, Region IDs, Region Names, URL to Corresponding Zillow Page (only for cities and neighborhoods), Latitudes and Longitudes, A region can be specified at various levels of the region hierarchy. An optional childtype parameter can also […]
  • Sift Science Score
    Sift Science Score API is a REST API that provides a risk score for users on an e-commerce website. The Sift score for users are assessed through JSON requests containing an API key and user id and provides a JSON response. This API allows for HTTP notifications to endpoints of the company's choice and up […]
  • Zillow Home Valuation GetComps
    The GetComps API allows for access to Zillow's database of recent sales for a specified property for comparision. Returned results aim to contain the address, Zillow property identifier, Zestimate for the comparable properties, and the principal property for which the comparisions are being retrieved. Date Updated: 2015-02-25 Tags: [field_primary_category], [field_secondary_categories]
  • Zillow Home Valuation GetChart
    The GetChart API allows developers to generate a URL for an image file that displays historical Zestimates for a specific property. The API uses the Zillow Property ID and a specified chart type: either percentage or dollar value change as an input. The historical data can be for the past 1 year, 5 years or […]
  • Sift Science Label
    Sift Science Label API is a REST API that labels bad user within the Sift Science system. This API allows the business user to label behavior, categorize events, and confirm or deny sift scores. The Sift Science system uses data to learn and predict bad behavior patterns. Labeling is made through a HTTP POST to […]
  • Yandex Dictonary
    Yandex, a European internet company, provides this dictionary API. Yandex Dictionary uses standard HTTP with access via XML or JSON interfaces. Use of the API requires an API Key which is available by request, for free, from Yandex. The app will return translations or word and phrase definitions. Yandex has a Terms of Use document […]
  • Happy Inspector
    The Happy Inspector API allows developers to access and integrate the functionality of Happy Inspector with other applications. Public documentation is not available; API access comes with account service. Happy Inspector is a platform for property and rental managers and management companies. Date Updated: 2015-02-25 Tags: [field_primary_category], [field_secondary_categories]
  • Yandex Speech Recognition
    This Speech Recognition API is provided by Yandex and requires an API Key, which are available by request. The API currently supports queries (general search requests) and maps (for accessing addresses) in Turkish and Russian languages. With interfaces for iOS, Android, and Windows Phone, this app offers developers the ability to integrate voice input interfaces […]
  • SigOpt
    The SigOpt API allows developers to access and integrate the functionality of SigOpt with other applications and websites. Some example API methods include creating and managing experiments, retrieving data and analytics, and managing accounts. SigOpt is an optimization platform that helps users optimize anything through fine-tuning experiments. Date Updated: 2015-02-23 Tags: [field_primary_category], [field_secondary_categories]