Discover and read the best of Twitter Threads about #Bigdata

Most recents (24)

Bayesian #MachineLearning Part 1 — Understanding Bayes Theorem from an Analytics Perspective: bit.ly/2KAJaHS #abdsc
—————
#BigData #DataScience #Statistics #StatisticalLiteracy #DataLiteracy #Algorithms
Bayesian #MachineLearning Part 3 — building Bayesian models & inference networks: bit.ly/2O2afGr #abdsc
—————
#BigData #DataScience #Statistics #StatisticalLiteracy #DataLiteracy #Algorithms
—————
+See also these books: amzn.to/30jbwuk
Read 5 tweets
Queridos amigos, hoy es sábado, sabadete, así que toca hilo de #privacidad y #eticadelosdatos. ¿Por qué es tan fácil para las empresas realizar perfiles de usuarios, y tan complicado para los usuarios acceder a esa información, incluso cuando es totalmente discriminatoria?
Vivimos en la era del #BigData Analytics, donde la toma de decisiones automatizada, el aprendizaje automático y las técnicas de creación de perfiles son cada vez más capaces de evaluar las vidas de las personas en función de su propia historia datos y predicciones.
Las técnicas automatizadas de toma de decisiones y creación de perfiles pueden ser perjudiciales para las personas porque la legislación actual no proporciona a los usuarios control sobre sus propios datos, y menos sobre las evaluaciones realizadas por los procesos automatizados.
Read 50 tweets
🆃🅴🅾🆁🅴🅼🅰 (de representación de Arnold-Kolmogorov)

Tiene que ver con #deeplearning, #machinelearning, AI, #bigdata y todas esas palabras cool (y es una excusa para hablar de Arnold).

Dedicado a @wsosaescudero y los fanáticos del libro verde.
🗣️ porfa RT así difundimos #TeRegaloUnTeorema
Dice así: toda función continua, aunque dependa de muchas variables se puede escribir como combinación de funciones (continuas) que dependen de una sola variable.
Read 13 tweets
@sverhulst is at #UNHQ today for the High Level Political Forum on Sustainable Development. In a few moments, he’ll be sharing GovLab’s experiences for a panel on how #bigdata can contribute to the follow-up on the #2030Agenda. Stay tuned! #HLPF2019
And we’re staring! Excited to hear how new data sources can inform official statistics for #SDGs
Director General Yusuf Murangwa discusses how Rwanda used #bigdata to improve its understand of the economy and agriculture.
Read 22 tweets
Thread: the public vs. procurement

<HUGE> development spending hides in plain sight.

Do you ignore it? Think someone else has it covered?

In LMICs upto 50% of all government spending is via <PUBLIC PROCUREMENT>

Myriad contracts: roads/bridges/hospitals/drugs....
1/24
(hand) Pump Primer:

India 1999: I worked in WATSAN. Over 1000 new handpump tubewells would replace contaminated shallow wells & rusting pumps that often stood in puddles. You paddled to collect drinking water, and the puddle seeped into the well.
2/24
The Good: The new pumps were installed on large concrete platforms, in turn mounted on a large buried concrete ring. This stopped water undermining the platform, which would crack and break up the concrete. Back to puddles and contamination.
3/24
Read 26 tweets
Sábado, sabadete, por lo que hoy toca hilo sobre #privacidad y #protecciondedatos o (des)protección de datos. Tema central de hoy: ¿Por qué estamos actualmente en una situación en la que la privacidad y la falta de transparencia se han convertido en cuestiones legales centrales?
Obviamente, se debe a un rápido desarrollo tecnológico, pero quizás sea útil para nuestra discusión sobre #transparencia, #privacidad y #perfiles.
para cavar un poco más profundo en el por qué, y no en el cómo. ¡Adelante hilo!
Dos hechos tecnológicos hicieron posible esta enorme transformación: 1) el desarrollo de hardware informático y 2) el desarrollo de un software que permitió que muchos ordenadores funcionaran como uno solo. ¿Qué supusieron estos grandes hitos para las organizaciones y empresas?
Read 35 tweets
Worth bearing in mind.

No one has deployed quantum encryption or quantum entanglement, yet, which are two of the easiest solutions if you have them.

But we need to augment civilian IT security dramatically, especially in strategic industries.
forbes.com/sites/zakdoffm…
And yes, the fact that many key breakthroughs apparently *haven't* filtered their way into cutting-edge industries is actually good news, given how much IP theft and general infiltration we seem to be dealing with.

Still odd how many things haven't surfaced yet.
Especially the blindingly obvious.

You really *didn't* know the malware could be tracked, and more importantly mapped, over years and even decades?

That your botnets were just a mountain of hard evidence?
Read 6 tweets
#DeterministicPhysics

Understand the laws and you understand the nature
#BeautyOfMathematics

No problem is ever unsolvable. You can also solve a problem by showing that no solution exists.
Laplace on #Determinism of the Universe - “Such an intelligence would embrace in the same formula the movements of the greatest bodies of the universe and those of the lightest atom; for it, nothing would be uncertain and the future, as the past, would be present to its eyes.”
Read 21 tweets
#China built a surveillance database on Americans from stolen/hacked data, such as the #OPM hack, and some US software developers helped them build & implement it.

I’d love to know who those (treasonous, IMO) developers were.
According to the insider, the database is very similar to China’s social credit scoring system that tracks every detail of Chinese citizen’s lives, upon which China Asians them a “freedom score.”
The software they used to build it out was originally a #BigData analytics program for #SmartCities, like the one Google wants to build in #Toronto.

forbes.com/sites/jilliand…
Read 3 tweets
Here is a summary of the @ASCO Minimal Common Oncology Data Elements (#mCODE) project ... in thread format
#ASCO19
1/ Problem:
#clinicaltrials continue to be the primary way to answer question in oncology but few patients enroll. Despite more data being stored in the #EHR, we can't easily combine data across systems as few are interoperable. This is a missed opportunity.
2/ Solution:
#mCODE is a newly proposed data standard that leverages and expands existing data standards to promote answering clinically meaningful oncology questions across medical records allowing us to learn from more patients
Read 8 tweets
Che bella la Master of Elusion #amazon che licenzia centinaia di ̷s̷c̷h̷i̷a̷v̷i̷ dipendenti ogni anno nei suoi magazzini in "sottoproduzione" (ossia che sforano al ribasso la rigida metrica stabilita da precisi algoritmi di Artificial Intelligence AI).

theverge.com/2019/4/25/1851…
Poi quando la Master of Elusion #amazon si accorge di un video che le reca un po' di prurito... succede che quel video, dopo anni di fiera militanza nel canale Tu(mi)Tu(r)bi dell'Unità, decide per partito preso di emigrare all'estero.

Read 37 tweets
I'm attending the #DataForGood panel organized by @SFU, moderated by @drjoyjohnson w/ Sheryl Groeneweg @ISED_CA, Richard Evans @StatCan_eng, and Bohdan Noysk @SFU

#AI #partnership #bigdata #cdnsci #scipol #cdnpoli
RE - risk assessment for data security is a daunting task ➡️ so it's important to be able to communicate what is known, what can be done. Govt needs to be forthcoming to build great relationships, to build trust.

BN - the best partnerships are built on strong relationships.
BN works in sensitive patient data, which is a fantastic resource. Researchers have been accessing this kind of data for decades because of good practise to anonymized and de-identify data.
Read 15 tweets
In July 1969 the Apollo 11 astronauts conducting several tasks during extravehicular activity (EVA) operations on the surface of the moon. The EVA lasted approximately 2.5 hours with all scientific activities being completed satisfactorily. #OneGiantLeap

#Apollo11 landed on the Moon on July 20, 1969.

That's #OneSmallStep for [a] man, #OneGiantLeap for mankind.

Read 54 tweets
BigData è passato dall'essere una semplice parola d'ordine a una realtà che sta iniziando ad avere un impatto significativo in tutti gli aspetti della nostra vita pubblica e privata. #BigData #BigDataAnalytics
Sia che si tratti di siti di social media che pubblicano annunci accuratamente mirati, sia che le città offrano servizi migliori ai loro residenti o sia che i governi non tutelino la privacy, l'impatto e il potenziale dei #BigData sta cominciando solo adesso ad essere intuito.
Gran parte dello sforzo profuso sembra essere focalizzato sulla componente analitica dei dati, come l'uso del giusto tipo di strumenti #BigData per estrarre le informazioni a partire dai dati e tracciare intuizioni fino ad allora sepolte in strati di dati.
Read 7 tweets
To understand why so many are highlighting the Chinese cameras in Leinster House, one must first recognise that China does not do business or espionage like the West. There's no separation between industry sectors the CCP deems strategically important.
The #IOT, #5G, #AI and #SmartCities sectors are integral to China. These sectors have billions of dollars pumped into them from Chinese government central and regional funds and are controlled centrally by the CCP, with the CCP placing key personnel into these companies.
One other area of interest to the Chinese government is #BigData - and boy do they have the opportunity to gather as much of this as possible thanks to the West. This data is used to forecast and predict actions, reactions of markets, industry and people - most importantly people
Read 27 tweets
#Bigdata vs Machine Learning vs Artificial Intelligence
By Irene Aldridge
☝️author High-Frequency Trading: A Practical Guide to #Algorithmic Strategies& #Trading Systems
☝️co-author Real-Time Risk: What Investors Should Know About #Fintech,#highfrequencytrading and FlashCrashes
1. ☝️🧐➿🔢 In traditional #statistics or #econometrics, researchers make assumptions about #data distributions ahead of the analysis
2. 🥇💪#machinelearning = the 1st discipline to apply #efficiency to #problemsolving brought by #computers & their enhanced computational power
3. 🤗➿🔢#ML scientists try to reduce #assumptions about the data as much as possible&let the data (&computers) decide what fits best.
4. ♾➡️🔤🌐#Datascience identifies core characteristics of the data, summarized by what has been known as #characteristic values (#eigenvalues)
Read 6 tweets
Data Needs A Home:

🧬The population of every DNA word is equivalent to the population of its reverse-complement DNA word in the SINGLE-STRANDED human genome

🧬Population count of 1.4 trillion DNA words

🧬Does anyone want a copy of this data before files are deleted?

#Genomics
Can You Use This #Data?

🧬Demonstrates the inherent inversion symmetry structure for all DNA elements in the single-stranded genome

🧬Demonstrates the absence of analog symmetries between reverse pairs and complement pairs of DNA elements

#BigData
🔂

👉cell.com/iscience/fullt…
Do You Know Anyone 🏷 Who Can Use This Data?

🧬The number of times every #DNA element (1-nt to 20-nt) occurs in the single-stranded genome

🧬This inversion symmetry structure is also maintained at the level of each individual #chromosome, except chromosome M (mitochondrial DNA)
Read 6 tweets
Last week, I did a talk with @WSU_COM PhD students about applying for #nonacademicjobs or what's often called #postac or #altac careers. My goal was to encourage students to explore their options, but also inject some reality in their sometimes starry-eyed discussions. 1/start
2/ Right off the bat, I wanted to dispel the notion that PhD jobs are a great fit for nonacademic jobs in industry, nonprofit or govt. sectors. There are a number of reasons for this: Nonacademic recruiters think you are too expensive or esoteric for their jobs...
3/ Continuing reasons why PhD doesn't easily translate to #AltAc: Faculty lack expertise or knowledge about nonacademic jobs (even if they want to support you), and graduate programs encourage deep scholarly knowledge and publication rather than broader public engagement.
Read 25 tweets
#bigdata is promising but not a panacea. We can miss obvious problems if we
rely only on big data. To enhance power of big data, we need to combine it with #thickdata - non-measurable, qualitative insights.

Sharing my IBM report, download 👇

deepblue.lib.umich.edu/handle/2027.42…
Limits of #bigdata 1⃣

Having lots of data can be overwhelming or have little utility
if the data are thin—that is, they lack meaning for users or fail to capture issues that matter most.
Limits of #bigdata 2⃣

Organizations may sometimes have big data on things that do not really
matter and yet lack data on issues that matter most. 🤔
Read 13 tweets
Perhaps the worst cover image for one of the worst books purporting to discuss #ethics and #bigdata.
“Big data technology has no value framework.” Bullshit. Big data is inherently techno-positivist, and largely enabled by infrastructures of surveillance lacking consent or respect for individual subjects.
I now need to write a book simply to respond to how fucking horrible this book is. @OReillyMedia, you really couldn’t find anyone better to tackle this topic?
Read 5 tweets
1/? THREAD: What can the world learn from Brazil’s struggle to #endslavery?

This week I’m visiting Brazil to consider that question.

Combining mobile inspection teams, social action & big data, Brazil’s efforts are amongst the most sophisticated in the world.
2/ >50000 modern slaves have been rescued by the Brazilian state since 1995.
On behalf of @UNUCPR @Delta87org & @FinComSlavery I'll meet with @undpbrazil @ilo @MPT_PGT @reporterb @experian & more, focusing on policy in agriculture & finance.

Learn more about slavery in Brazil & government efforts here: delta87.org/dashboards/bra…

Follow along - & send ?s!
Read 68 tweets
[#SmartFuture][#February27]
─The internet will not scale to meet the demands of #AI, #IoT and video streaming without active participation from service providers.
─A lightweight autonomous underwater vehicle for a maximum depth of 200m. The Sparus II is capable of hovering and has an easily configurable payload area and an open software architecture.
Read 15 tweets
Hello Twitter friends, which do you think is a better cover? Other ideas for a cover design? #BigData

This is 4 forthcoming IBM report ➡️
Integrating Big Data & Thick Data to Transform Services Delivery

Abstract 👇
P1: Big data holds great promise for improving public services delivery & innovation in government, but it is not a panacea. Having lots of data can be overwhelming or even useless for agencies, if the data are “thin”— that is, they lack meaning for users.
P2: Governments can greatly enhance the value of big data by combining it with “thick” data—rich qualitative information about users, such as their values, biases, and previously unexpressed needs, which are obtained by interacting with or observing them in their daily lives.
Read 9 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!