sync ; echo 3 | sudo tee /proc/sys/vm/drop_caches

Remote terminal application that allows roaming, supports intermittent connectivity, and provides intelligent local echo and line editing of user keystrokes.

Mosh is a replacement for SSH. It's more robust and responsive, especially over Wi-Fi, cellular, and long-distance links.

Mosh is free software, available for GNU/Linux, FreeBSD, and Mac OS X.

Learning To Classify Text Using Support Vector Machines gives a complete and detailed description of the SVM approach to learning text classifiers, including training algorithms, transductive text classification, efficient performance estimation, and a statistical learning model of text classification. In addition, it includes an overview of the field of text classification, making it self-contained even for newcomers to the field. This book gives a concise introduction to SVMs for pattern recognition, and it includes a detailed description of how to formulate text-classification tasks for machine learning.

Texts semantically annotated with WordNet 1.6 senses (created at Princeton University), and automatically mapped to WordNet 1.7, WordNet 1.7.1, WordNet 2.0, WordNet 2.1, WordNet 3.0

ries (or RIES, an acronym for RILYBOT Inverse Equation Solver) takes any number and produces a list of equations that approximately solve to that number, like the following example

Our dataset consists of 11881 annotated lines from almost 400 email messages drawn at random from the Enron email corpus. We use the database dump of the Enron corpus (219Mb) released by Andrew Fiore and Jeff Heer. This version of the corpus has been processed to remove duplicate messages and to normalise sender and recipient names, resulting in just over 250,000 email messages. No attachments are included. Our annotations are made by a single annotator.

This is a partial list of work related to the AAAI-11 Workshop on Analyzing Microtext.

line-by-line analysis of memory usage for Python code

The roots of ILL can be traced back to De Saussure (1916) and Bloomfield (1933): language tasks can be learned and performed by employing analogy and induction on relations between language elements. Chomsky strongly criticised the bluntness of the analogy/induction approach, capitalising on its inability to capture relations involving meaning. In ILL, the pre-Chomskyan ideas on analogy and induction are implemented on present-day computer technology, using general-purpose inductive-learning tools developed in machine learning to explore the range of language tasks that can be learned successfully. After giving the historical background and an introducting into supervised inductive machine learning, the course will provide an overview of methods, techniques, and empirical results showing that ILL is successful in morpho-phonology, and (more suprisingly) succesful in several higher-level language tasks (e.g., POS tagging, PP-attachment).

Resources for Chinese grammar have been scattered, incomprehensible, and hidden behind paywalls for way too long. Wikipedia has shown us a better way, and we have taken up the challenge of making Chinese grammar learner-friendly and accessible to all. We've already got 506 articles online, and we're growing!

I made this little program so you can view and link to a whole conversation from Twitter in context. Here's an example! Want one for your conversation? Just enter the URL of an individual tweet (it should contain a long number):

SVMsequel is a complete environment for training and use support vector machines. Some familiarity with kernel methods will be helpful (see here for class notes I've used for using SVMs for natural language processing if you need a refresher.

Numerous linguists have praised Hangul for its featural design, describing it as "remarkable", "the most perfect phonetic system devised", and "brilliant, so deliberately does it fit the language like a glove."[23] The principal reason Hangul has attracted this praise is its partially featural design: The shapes of the letters are related to the features of the sounds they represent: The letters for consonants pronounced in the same place in the mouth are built on the same underlying shape. In addition, vowels are made from vertical or horizontal lines so that they are easily distinguishable from consonants.

Latino sine flexione ("Latin without inflections"), or Peano’s Interlingua (abbreviated as IL), is an international auxiliary language invented by the Italian mathematician Giuseppe Peano (1858–1932) in 1903. It is a simplified version of Latin, and retains its vocabulary. It was published in the journal Revue de Mathématiques, in an article entitled De Latino Sine Flexione, Lingua Auxiliare Internationale,[1] which explained the reason for its creation. The article argued that other auxiliary languages were unnecessary, since Latin was already established as the world’s international language. The article was written in classical Latin, but it gradually dropped its inflections until there were none.

A drop placed at the surface of the same liquid coalesces within a few tenths of seconds. Vibrating the bath of liquid on which the drop is placed can inhibit this process. The drop will then be able to bounce at the surface of the liquid for an unlimited time. In this thesis, we use liquids of medium to low viscosity. A bouncing drop then emits a wave at the surface of the liquid at each bounce. Those drops spontaneously organize themselves in bounded states or in clusters. Just below the Faraday instability threshold, a remarkable phenomenon occurs when the drop undergoes a drift bifurcation and starts moving horizontally at the surface of the liquid, acquiring a constant horizontal velocity. We call such drops walkers. We have studied this transition from a steady bouncing drop to a walker and described it theoretically. A walker never collides directly with one of the cell's walls but, via its own waves and the waves emitted at the boundaries, is repelled and undergoes a reflection. Thus in certain situations the drop can have a billiard-like motion in the cell. We have also observed the various collisions (always via their waves) of several walkers moving across the cell. The attractive collision of two walkers leads to the orbiting motion of the two drops. The size of the orbits can take a series of discrete values, which can be explained by the interaction of the drops via the interferences created by their associated waves. We also discuss the differences and similarities between these new objects and localized structures observed in various 2D dissipative systems such as oscillons in fluids and granular materials or cavity solitons in optics.

The Haskell Lightweight Virtual Machine (or, informally, the HaLVM) is a port of the  GHC runtime system for Haskell to barebones  Xen. This means that Haskell programs written for the HaLVM run natively on Xen, without any intervening operating system, which allows them to boot quickly and use very little space.

Gensim is a free Python framework designed to automatically extract semantic topics from documents, as efficiently (computer-wise) and painlessly (human-wise) as possible.

Gensim aims at processing raw, unstructured digital texts (“plain text”). The algorithms in gensim, such as Latent Semantic Analysis, Latent Dirichlet Allocation or Random Projections, discover semantic structure of documents, by examining word statistical co-occurrence patterns within a corpus of training documents. These algorithms are unsupervised, which means no human input is necessary – you only need a corpus of plain text documents.

Once these statistical patterns are found, any plain text documents can be succinctly expressed in the new, semantic representation, and queried for topical similarity against other documents.

SimString is a simple library for fast approximate string retrieval. Approximate string retrieval finds strings in a database whose similarity with a query string is no smaller than a threshold. Finding not only identical but similar strings, approximate string retrieval has various applications including spelling correction, flexible dictionary matching, duplicate detection, and record linkage.

In this article we shall demonstrate how simple shell scripts can be used to implement sets, providing one line recipes for set creation, set union, set intersection and more. Having explored the power of the Unix shell we’ll consider its limitations, before finally discussing the more general lessons we can learn from the Unix tools.

YourQA Corpora

These are the TREC'01 non-factoid answer corpora used in the following papers

Can a machine be taught to determine the sentiment of a Twitter message about weather?  With the data from over 1 million crowd sourced human judgements the goal was to use this data to train a predictive model and use this machine learning system to make judgements.  Below are the highlights from the research and development of a machine learning model in the cloud that predicts the sentiment of text regarding the weather.  The following are the major technologies used in this research:  Google Prediction API, CrowdFlower, Twitter,  Google Maps.

We interpret the SVM algorithm as the maximum a posteriori solution to a Bayesian inference problem. It is then natural to select hyperparameters to maximize the evidence, i.e. the overall likelihood of the observed data. The key advantage is that the evidence is a continuous function of the hyperparameters, and so can be optimized by e.g. gradient ascent. We have tested this method on a number of standard data sets and found very encouraging results. For details and background references, see the papers on SVMs in my publications list.

This article is an attempt to better understand how the front-end of Python works. Just reading documentation and source code may be a bit boring, so I’m taking a hands-on approach here: I’m going to add an until statement to Python.

These are mathematicians frequently encountered by undergraduate students. For a more complete listing, A History of Mathematics: An Introduction , by Victor Katz (Harper Collins) has a wonderful pronunciation guide as part of the index. Interested persons are urged to consult this text.

On-line corpora with concordancers
Free corpora for downloading
Subscription corpora and tools
On-line corpora of web texts
Transcribed spoken English (some audio)
Web as Corpus (corpora on the fly)
Concordancers and Tools

Much of my research is in topic models, which are a suite of algorithms to uncover the hidden thematic structure of a collection of documents. These algorithms help us develop new ways to search, browse and summarize large archives of texts.

Below, you will find links to introductory materials, corpus browsers based on topic models, and open source software (from my research group) for topic modeling.

setbeamertemplate{footline}[frame number]

I've bundled the whole pdfmarks-generation bit into a script, pdf-merge.py, which generates the pdfmark file and runs Ghostscript automatically. Think of it as a bookmark-preserving version of pdftk's cat. The script uses pdftk internally to extract bookmark information from the source PDFs.

It's probably muted. Open the terminal and start up the alsamixer:

Code:
alsamixer -c0 -Vcapture
Select the recording input device - choose between Mic and Front Mic to get sound in Skype. Use up and down arrows...

The table below lists English words that have irregular plurals.

This is a list of irregular verbs in the English language. The citation form (the infinitive) comes first (with a link to the Wiktionary article on the verb), together with the present tense forms when they are different, then the preterite or simple past, and finally the past participle. The right hand column notes whether they are weak or strong and whether they belong to a subclass, and links to discussions elsewhere. Typical irregularities in weak verbs are the assimilation of dentals (bended → bent) and vowel reduction (*keeped → kept).

This is a list of English copulae. Because many of these copulative verbs may be used non-copulatively, examples are provided.

Catenative verbs are verbs which can be followed directly by another verb — variously in the to-infinitive, bare infinitive or present participle/gerund forms. For example He deserves to win the cup, where deserve is a catenative verb which can be followed directly by another verb, in this case in the to-infinitive form.
Most of these verbs demand that the following verb be in one or the other form only. A few can take both forms, but sometimes there is a difference in meaning.
They are called catenative from their ability to form chains. We promised to agree to try practicing playing tennis more often.

This set of reference grammars has been designed for advanced-level language users and linguists to compare semantic categories across languages. Each grammar also provides background information about the language and its speakers. In some cases the author has included a topic which provides greater illumination of the language (e.g. tongue twisters, slang/profanity, or a set of exemplary texts). These sections are indicated by a shaded background in the left-side navigation pane.

Eventually this will be a glossary of symbol names in alphabetical order with notes. Some of these names are specific to particular packages or language pairs, as not all languages have the same grammatical features (most don't have spatial distinction in articles for example).

The Leipzig Glossing Rules have been developed jointly by the Department of
Linguistics of the Max Planck Institute for Evolutionary Anthropology
(Bernard Comrie, Martin Haspelmath) and by the Department of Linguistics
of the University of Leipzig (Balthasar Bickel). They consist of ten rules for the
"syntax" and "semantics" of interlinear glosses, and an appendix with a
proposed "lexicon" of abbreviated category labels. The rules cover a large part
of linguists' needs in glossing texts, but most authors will feel the need to add
(or modify) certain conventions (especially category labels). Still, it will be
useful to have a standard set of conventions that linguists can refer to, and the
Leipzig Rules are proposed as such to the community of linguists. The Rules
are intended to reflect common usage, and only very few (mostly optional)
innovations are proposed.

The World Atlas of Language Structures (WALS) is a large database of structural (phonological, grammatical, lexical) properties of languages gathered from descriptive materials (such as reference grammars) by a team of 55 authors (many of them the leading authorities on the subject).

At its core, Tatoeba is a large database of example sentences translated into several languages. But as a whole, it is much more than that.

The Wiktionary Look Up tool is a javascript which allows readers to simply double-click on a term to receive a pop-up with the top definition for the term, without leaving the page they are reading. It's particularly useful when visiting a site not in your first language, but great whenever you run across a term you're not quite certain about.

GFE is a free web-based function graphing tool that allows you to plot up to three functions on the same set of axes. In the functions you can refer to up to four independent variables that are controlled by sliders. This allows you to easily see the effect of changes since the graphs change in real time as you drag the sliders.

An apparently new speech disorder a linguistics department our correspondent visited was affected by has appeared. Those affected our correspondent a local grad student called could hardly understand apparently still speak fluently. The cause experts the LSA sent investigate remains elusive. Frighteningly, linguists linguists linguists sent examined are highly contagious. Physicians neurologists psychologists other linguists called for help called for help called for help didn’t help either. The disorder experts reporters SpecGram sent consulted investigated apparently is a case of pathological center embedding.

In mathematics, a Golomb ruler is a set of marks at integer positions along an imaginary ruler such that no two pairs of marks are the same distance apart.

Time loop logic, coined by the roboticist and futurist Hans Moravec,[11] is the name of a hypothetical system of computation that exploits the Novikov self-consistency principle to compute answers much faster than possible with the standard model of computational complexity using Turing machines. In this system, a computer sends a result of a computation backwards through time and relies upon the self-consistency principle to force the sent result to be correct.

The conjecture has been checked by computer for all starting values up to 20 × 258 ≈ 5.764×1018.[9] All initial values tested so far eventually end in the repeating cycle {4,2,1}, which has only three terms. It is also known that {4,2,1} is the only repeating cycle possible with fewer than 35400 terms.[10]
Such computer evidence is not a proof that the conjecture is true. As shown in the cases of the Pólya conjecture, the Mertens conjecture and the Skewes' number, sometimes a conjecture's only counterexamples are found when using very large numbers. Since sequentially examining all natural numbers is a process which can never be completed, such an approach can never demonstrate that the conjecture is true, merely that no counterexamples have yet been discovered.

General instructions to rebuild a Fedora source rpm.

A concordance is an index of all the words in a text, along with their contexts. A concordance-like functionality -- an alphabetical listing of all words in a text and their frequency -- can be made fairly easily with some basic shell tools: tr, sort, and uniq.

In computer science and discrete mathematics, an inversion in a sequence of numbers is a pair of numbers in the sequence that are "out of order" with respect to an ascending or descending order.

QEMU can launch Linux processes compiled for one CPU on another CPU, translating syscalls on the fly.

The binfmt-misc module in the kernel makes it possible to execute binaries of foreign arches under linux. Qemu can use this fact to enable several architecture specific execution environments in userspace without the need to run a kernel of the target architecture. ARM EABI support is included in the 0.11.x release used in ubuntu karmic today.