Bootstrapping methods [8][9][12]

Bootstrapping methods use the labeled data to train an initial model and then use that model to label the unlabeled data and re-train a new model.

Graph regularization methods [4]

Graph regularization methods build models based on a graph on instances, where edges in the graph indicate…

construct IsA network(probase[1]):

  1. iterative extraction: get super and sub concept pair ( isA, isAttributeOf, etc)

a. syntactic extraction: extract super and sub concepts candidates

b. super concept detection: detect super concept for the sentence( observation 1)

c. sub concept detection: detect sub concept for the sentence(observation 2)

2. taxonomy construction…

To continue from Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500 — part1: data preparation and model

general results

the above figure shows Daily performance metrics for long-short portfolios of different sizes: Mean return, standard deviation, and directional accuracy from December 1992 until October 2015. …

contribution of this paper:

  1. unique in deploying three state-of-the-art machine learning techniques and their simple ensemble on a large and liquid stock universe
  2. It reveals that ensemble returns only partially load on systematic sources of risk, are robust in the light of transaction costs, and deteriorate over time — presumably…

syntax-agnostic dependency based srl

notice: syntax agnostic here means tested on ood works better than other methods. Because syntactic parsers are unreliable on out-of-domain data, so standard (i.e. syntactically-informed) SRL models are hindered when tested in this setting(ood).


  1. first identify predicates and disambiguate them
  2. then, for each predicate, we re-encode the sentence with…

this article is mainly a summarization of Yasemin Altun’s presentation in May 2014 on how google applies text summarization.

text summarization is highly related to google knowledge graph project:

entities description within red circle use text summarization from wiki to give a one sentence description of the entity.

Wenchen's ai fantasy

nerd by train, leading purposeful life, trying to make a big impact. self-taught entrepreneur to be.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store