Come to the online (and free) Apachecon and watch us talk about the core of the Universal Recommender -- the Correlated Cross-Occurrence (CCO) algorithm, implemented in Apache Mahout.
apachecon.com/acah2020/track…
1
Apachecon goes virtual and the Universal Recommender will be there. Check out the Mahout track: apachecon.com/acna2020/track…
Use of Deep Learning is becoming very difficult if you don't have huge data. We've said it before. Here is a way to extract features from giant models that work with traditional ML. I think of is a condensing the information into more easily digests bits. Great read.
Now a good explanation for the "never enough data to make Deep Learning perform" effect and how to turn NNs trained on truly huge data into feature extractors for use with more traditional ML. Highly recommended article! basilica.ai/blog/the-unreaso…
The Universal Recommender v0.7.3 released. Now with cross-recommendations; "viewed this bought that" Use it to help a user find the target of a browsing or search session. groups.google.com/forum/#!to…
PredictionIO graduates to a top level project at Apache. The Mahout based Universal Recommender runs on it. globenewswire.com/news-relea…
A personal story of obsession and the gravitational attraction of Apache thru Mahout and PredictionIO blogs.apache.org/foundation/…
#apachemahout now runs on GPUs, this is one example of what is next. extremetech.com/extreme/2491… via @ExtremeTech
1
ActionML retweeted
Commodity clusters may give way to massively parallel multi-core chips and speculative execution
phy.so/418284964 via @physorg_com
Multi-domain predictive AI or how to make one thing predict another Mahout's CCO in a Recommender developer.ibm.com/dwblog/201… via developerworks
ActionML retweeted
What will the dressing rooms of the #future look like? developer.ibm.com/dwblog/201… #nearFuture @ApacheMahout #IoT @developerWorks
IBM's Watson may get most of the headlines but Ted talkster Lisa Seacat calls out the algorithm inside the Universal Recommender here. nitter.cf/occam/status/867067292…
Ted talker Lisa Seacat talks about Apache Mahout and CCO's mashup with the IoT in next gen commerce. developer.ibm.com/dwblog/201…
1
ActionML retweeted
Great talk by Andrew Musselman! @mlconfsea closing out the first half!
The Universal Recommender and the landscape of PredictivAI with the CCO algoithm at the GTC GPU conference. #GTC17 nitter.cf/suneelmarthi/status/86…
ActionML retweeted
Meet PredictionIO, now in development @ the Apache Incubator
incubator.apache.org/project… #Apache #OpenSource #MachineLearning #server @ApachePIO
ActionML's Universal Recommender at scale nitter.cf/ApacheMahout/status/78…
Come out to see @occam talk about scaling recommenders in production tonight! meetup.com/Seattle-Scalabili…
1
ActionML retweeted
Describes part of the Universal Recommender with PredictionIO: actionml.com/docs/ur nitter.cf/suneelmarthi/status/77…
Good overview of @ApacheMahout Co-Occurrence Recommenders by @mapr @caroljmcdonald ln.is/www.mapr.com/blog/SQh5… cc:@occam
Some notes on debugging your Universal Recommender model: actionml.com/docs/ur_elastic… #PredictionIO #MachineLearning #Recommenders
1














