]> Understanding LSTM Networks -- colah's blog 2015-10-16T14:24:57Z 2015-10-16 2015-10-13T10:23:07Z Web design tool, CMS, and hosting platform | Webflow 2015-10-13 Tokenization, POS Tagging, removing punctuation, stopwords, stemming, frequency distributions, Collocations, Bigrams, Trigrams, chunking, Splitting Training Sets + Test Sets, classifiers & scikit-learn, Cross Validating Classifiers, pipelines for classifiers Python NLP - NLTK and scikit-learn 2015-10-21 2015-10-21T18:36:28Z 2015-10-10 2015-10-10T20:47:50Z Histography - Timeline of History 2015-10-13 2015-10-13T17:05:51Z Commons CSV 2015-10-10 2015-10-10T20:57:09Z Strapdown.js - Instant and elegant Markdown documents 2015-10-21T10:08:08Z scikit-learn tutorial about analysing a collection of labelled text documents : - load the file contents and the categories - extract feature vectors (count, tf, tf-idf) - train a linear model to perform categorization - use a grid search strategy (to find a good configuration of both the feature extraction components and the classifier) Working With Text Data — scikit-learn documentation 2015-10-21 MarkdownImplementations - Markdown Community Group 2015-10-11T09:42:26Z 2015-10-11 A Simple Markdown Spotlight Importer - Hiltmon 2015-10-14T13:32:14Z 2015-10-14 pour faire en sorte que spotlight indexe bien les fichiers md 2015-10-21T16:56:28Z How to preprocess labeled data for use with SciKit-Learn - Quora 2015-10-21 A relatively easy way (though not the only one) is to use Natural Language Toolkit (NLTK)'s scikitlearn module in the classify Package co-operating.systems 2015-10-18T19:51:16Z 2015-10-18 Faut-il prendre l’effondrement au sérieux ? | InternetActu 2015-10-18T19:32:36Z 2015-10-18 Percy Liang Knowledge graphs often have missing facts (edges) which disrupts path queries. Recent models for knowledge base completion impute missing facts by embedding knowledge graphs in vector spaces. We show that these models can be recursively applied to answer path queries, but that they suffer from cascading errors. This motivates a new "compositional" training objective, which dramatically improves all models' ability to answer path queries, in some cases more than doubling accuracy. John Miller Traversing Knowledge Graphs in Vector Space 2015-08-19T05:16:24Z [1506.01094] Traversing Knowledge Graphs in Vector Space 1506.01094 2015-10-31 Kelvin Guu Kelvin Guu 2015-06-03T00:38:25Z 2015-10-31T00:11:12Z Path queries on a knowledge graph can be used to answer compositional questions such as "What languages are spoken by people living in Lisbon?". However, knowledge graphs often have missing facts (edges) which disrupts path queries. Recent models for knowledge base completion impute missing facts by embedding knowledge graphs in vector spaces. We show that these models can be recursively applied to answer path queries, but that they suffer from cascading errors. This motivates a new "compositional" training objective, which dramatically improves all models' ability to answer path queries, in some cases more than doubling accuracy. On a standard knowledge base completion task, we also demonstrate that compositional training acts as a novel form of structural regularization, reliably improving performance across all base models (reducing errors by up to 43%) and achieving new state-of-the-art results. Weebly Website Builder: Create a Free Website, Store or Blog 2015-10-13T10:18:14Z 2015-10-13 2015-10-11 Vue.js is a library for building modern web interfaces. It provides data-reactive components with a simple and flexible API. vue.js 2015-10-11T09:46:57Z 2015-10-13 2015-10-13T10:26:55Z Après l’attentat d’Ankara, la Turquie au bord du gouffre scikit-learn: machine learning in Python 2015-10-16 2015-10-16T00:17:32Z NLTK (Natural Language Toolkit) - home 2015-10-21T18:38:33Z 2015-10-21 2015-10-31T10:16:58Z 2015-10-31 Le fact-checking peut-il s’automatiser ? | J'ai du bon data Using CORS - HTML5 Rocks 2015-10-13T11:53:30Z 2015-10-13 2015-10-04T11:40:05Z MD-LD MD-LD extends Markdown's reference link syntax to allow easy authoring of structured data. 2015-10-04 Pipeline and FeatureUnion: combining estimators — scikit-learn documentation 2015-10-21 2015-10-21T23:04:49Z Sample code: <a href="http://scikit-learn.org/stable/auto_examples/hetero_feature_union.html">Feature Union with Heterogeneous Data Sources</a> A Tower of Molten Salt Will Deliver Solar Power After Sunset - IEEE Spectrum 2015-10-24T22:28:00Z 2015-10-24 For the first time, solar thermal can compete with natural gas during nighttime peak demand Conférence : La blockchain au delà du Bitcoin 2015-10-03T12:43:10Z 2015-10-03 Robotics, automation play a big role in Gartner's top 10 predictions | Computerworld 2015-10-24T22:25:41Z 2015-10-24 Machines will replace more writers, and by 2018, 2 million workers will be required to wear health monitors Colah's blog 2015-10-16 2015-10-16T16:36:29Z The most important thing going on in the world today is something we almost never cover: a rapid decline in poverty, illiteracy and disease. The Most Important Thing, and It’s Almost a Secret - The New York Times 2015-10-13T10:09:33Z 2015-10-13 cars, street lights, and even buildings will behave in the same vaguely suspicious way 2015-10-03 2015-10-03T10:16:35Z The price of the Internet of Things will be a vague dread of a malicious world Feature extraction — scikit-learn documentation 2015-10-21T17:12:20Z 2015-10-21 Deep Learning, NLP, and Representations - colah's blog 2015-10-16T11:40:15Z 2015-10-16 2015-10-20T11:00:30Z Challenges of the email domain for text classification JD Brutlag, C Meek - ICML, 2000 - research.microsoft.com<br/> Interactive classification of email into a userdefined hierarchy of folders is a natural domain for application of text classification methods. This domain presents several challenges. First, the user's changing mailfiling habits mandate classification technology ... 2015-10-20 2015-10-12 2015-10-12T19:21:09Z html - Cross-reference (named anchor) in markdown - Stack Overflow Scikit Learn - Can you create your own dataset/data files to use with the code they use in the tutorials? - Quora 2015-10-21T16:48:54Z 2015-10-21 The great chain of being sure about things | The Economist 2015-10-31T23:53:56Z 2015-10-31 Obstacles on the Path to AI 2015-10-24T00:49:43Z 2015-10-24 2015-10-21 2015-10-21T18:43:13Z The Glowing Python: Combining Scikit-Learn and NTLK Turtles all the way down | Ruben Verborgh APIs are more than just data: context and controls also belong in the message. 2015-10-07 2015-10-07T22:42:37Z 2015-10-19T10:51:16Z 2015-10-19 scikit learn: machine learning map 2015-10-19 2015-10-19T12:18:52Z What is NumPy? 2015-10-03T10:09:21Z 2015-10-03 What US intelligence predicted about Africa today 15 years ago, and how terribly wrong they were | Mail & Guardian Africa (Mobile edition) comparison of several feature selection and extraction methods in the frame of email classification Highly discriminative statistical features for email classification 2015-10-20 2015-10-20T11:30:52Z 2015-10-11T09:51:45Z evilstreak/markdown-js 2015-10-11 2015-10-15 2015-10-15T10:29:50Z rdfjs/rdfjs.org python - scikit-learn, add features to a vectorized set of documents - Stack Overflow 2015-10-21 2015-10-21T23:14:21Z