Business intelligence is now taking off with companies like Endeca and NetSuite fighting for market share along with more established companies such as IBM and Microsoft. The idea of real time integration of all types of business data so that one can make more informed predictions and decisions is simple and obvious. Yet pulling this off in a business environment, with enough data format acronyms to make a reverse engineer cringe, is complex.
Continue Reading...Posts Tagged ‘learning’
Business Intelligence must be Automated
Sunday, December 6th, 2009Literature Review 2008 – 2009
Friday, November 6th, 2009The research we do at Helioid involves a lot of reading. With some notes and summaries included, here is a list of the literature we’ve focused on from 2008 to 2009:
Machine Learning
G. Lebanon, Y. Mao, and J. Dillon. The Locally Weighted Bag of Words Framework for Document Representation. Journal of Machine Learning Research 8 (Oct):2405-2441, 2007.
The Intentional Web
Thursday, February 26th, 2009The majority of the time one browses or searches the web there is a goal in mind. Find the location of a coffee shop, learn more about cloud computing, see if there are any interesting new movies, be distracted and procrastinate. Each of these instances of web use has objectives and implicitly defines a success predicate. When one (the agent) interacts with the web, a computer or simply information (the system), that systems knowledge or discovery of an explicit representation of the agent’s objectives, and the success predicates for these objectives, greatly enhances its capability to assist the agent in accomplishing its objectives.
The intentional web is a community of agents interacting with each other to accomplish their goals and increase their fitness.
Composing Inverse Functions to Measure Model Fitness
Friday, December 5th, 2008This articles concerns a method for evaluating the fitness of content topic models and document topic models based on the dissonance between a set of documents and the set of documents generated by composing inverse functions and applying them to the original set of documents. A document generating function is applied to a topic generating function that is applied to the original set of documents. In order to compare topics, one can look at the original set of topics compared to set of topics generated by apply a topic generating function to the documents generated by applying a document generating function to the original set of topics.
Continue Reading...Deficiencies of non-Non-linear Learning
Tuesday, November 25th, 2008Google’s failure to present the user with relevant results stems from an inability to capture user feedback and implement a method for users to provide feedback. The failure in part stems from the assumption that a user wants to go through their search results in a linear fashion.
Here are a couple ways in which Helioid’s non-linear approach to searching helps solve this problem.