jisaacso

10 ms·

31.
by jisaacso 10y ago
Take a look at the challenge homepage, http://physionet.org/challenge/2016/ tl;dr ECG is much more accurate but requires a careful, controlled environment. If it's possible to accurately predict heart abnorma
32.
by jisaacso 10y ago
All of the data can be found on the physionet site http://physionet.org/physiobank/database/challenge/2016/
33.
by jisaacso 10y ago
Hey thanks for the great questions! (1) The NN uses two convolutional units and a fully connected softmax layer. Relative to Inception V3 or highway networks , this is _not_ a very deep architecture. I was looking for a balance between accu
34.
by jisaacso 10y ago
The ML community has techniques for building methods to address low signal to noise, class imbalance, noisy labels and smaller sample sizes. I definitely agree, all of these are issues in medical datasets. Part of the exciting challenge at
35.
by jisaacso 10y ago
Thanks! I definitely agree that collecting these signals is difficult to scale.
36.
by jisaacso 10y ago
I definitely agree! The challenge is collecting a large number of heart recordings from mobile phones, along with professionally diagnosed abnormalities. That's one reason this physionet dataset is so valuable.
37.
DeepHeart: A Neural Network for Predicting Cardiac Health(github.com)
149 points by jisaacso 10y ago|21 comments
38.
URX drives mobile commerce through relevant product ads(blog.urx.com)
5 points by jisaacso 11y ago|0 comments
39.
by jisaacso 11y ago
Check out http://avisingh599.github.io/deeplearning/visual-qa/
40.
by jisaacso 11y ago
Recurrent Neural Networks. They're so cool. Used in everything from creating chat bots to describing pictures to machine translation. Bengio and Goodfellow have a preprint book available https://goodfeli.github.io/dlboo
41.
by jisaacso 11y ago
Really cool, thank! The `types` seem to be a good way to add context if you know, a priori, the type you're looking for. Google definitely fuses user data into their knowledge graph. This is seen in Freebase's `g.` identifier [1].
42.
by jisaacso 11y ago
Completely agree! I'm curious if the `query` parameter in the API performs well on long queries (with context) or if it needs to be focused to a single entity's name
43.
by jisaacso 11y ago
I'm curious how the knowledge graph API performs disambiguation without any context. E.g., if you search for `Apple` will it return the company or the fruit?
44.
by jisaacso 11y ago
Google's knowledge graph was in part based on Freebase, a large open knowledge base written by Metaweb. Google acquired Metaweb, continued to grow it's triple extractors [1] and eventually shut write access to the graph. Wikidata
45.
Structure from chaos: Using DOM structure to recommend mobile actions(blog.urx.com)
8 points by jisaacso 11y ago|0 comments
46.
by jisaacso 11y ago
DBpedia is a great, structured representation of Wikipedia. Agreed, there are many challenges, NLP being one. I think you'll like our other blog posts http://blog.urx.com/urx-blog/2015/7/28/named-ent
47.
by jisaacso 11y ago
Great example! This is the exact problem URX is solving: empowering content providers to present their products within meaningful contexts.
48.
How We Derive Context from Big Data(blog.urx.com)
25 points by jisaacso 11y ago|4 comments
49.
by jisaacso 11y ago
Opentable gave a great presentation on this not too long ago, check it out http://www.slideshare.net/SudeepDasPhD/recsys-2015-making-me...
50.
by jisaacso 11y ago
Great suggestion. We've used LDA for topic modeling in the past. I'm a big fan of running word2vec and clustering.
51.
by jisaacso 11y ago
Totally agree! Our goal was to work with Insight Data Science to build a lightweight, simple keyword extractor in 3 weeks.
52.
Automatic Keyword Extraction from Text(blog.urx.com)
64 points by jisaacso 11y ago|7 comments
53.
Show HN: Deep Learning in the Cloud(github.com)
1 points by jisaacso 11y ago|0 comments
54.
by jisaacso 12y ago
Author here, The article discusses applying PageRank to help decide how a web crawler should discover new content. While this article does not touch on how URX evaluates search quality, I can promise it's a metric we care deeply about.
55.
The Science of Crawl: Improving on PageRank(blog.urx.com)
35 points by jisaacso 12y ago|7 comments
56.
by jisaacso 12y ago
Hi Abraham, could you describe some of the differences between Atomwise's approach and D.E. Shaw research's approach to computational identification of novel pharmaceuticals? Thanks!
57.
Elasticsearch: A Simple Tutorial(github.com)
1 points by jisaacso 12y ago|0 comments
58.
The Science of Crawl, Part 2: Content Freshness(blog.urx.com)
27 points by jisaacso 12y ago|1 comments
59.
KDD Retro: Google Knowledge Vault and Topic Modeling(blog.urx.com)
26 points by jisaacso 12y ago|0 comments
60.
by jisaacso 12y ago
Thanks for the reference. It looks like SequenceMatch is "cubic time in the worst case and quadratic time in the expected case". Did you notice any performance issues as kouio scaled?

More ›