Senzari continue the evolution of their MusicGraph platform with the introduction of MusicGraph.ai at the Graphlab Conference 2014. MusicGraph.ai is both a dashboard for the music data and recommendations provided by MusicGraph and also allows for deeper mining and exploration with the addition of analytics and intelligence algorithms. MusicGraph.ai is intended to make algorithmic analysis as accessible as MusicGraph makes music data.
At the time Senzari COO Demian Bellumio discussed the longer terms plans for MusicGraph to not only broaden the availability of music data and recommendations but to develop MusicGraph as a "platform for connecting data and services that users might one day improve with their own algorithmic contributions."
Yesterday at GraphLab Conference 2014 Senzari unveiled MusicGraph.ai which provides a dashboard and analytics for the MusicGraph API and adds additional analytics and intelligence capabilities for big data crunching:
"MusicGraph.ai seamlessly integrates leading Big Data technologies, including Titan with Cassandra and Elasticsearch for the distributed graph database system, Hadoop and Spark for parallel processing, and GraphLab and GraphX as algorithmic engines."
"The algorithms will be divided in two distinct sets, Analytics and Intelligence. Analytics will initially include a set of popular 'graph analytics' algorithms such as PageRank, connected components, and triangle counting. Intelligence will incorporate powerful 'machine learning' algorithms, such as collaborative filtering, k-means, logistic regression, and support vector machines (SVMs)."
"In terms of real-world use cases for MusicGraph.ai, data scientists can leverage it to enhance or develop music recommendation services, identify data anomalies, and perform community detection within their networks, among other applications, all in an automated and highly scalable way."
Demian Bellumio states:
"“We believe MusicGraph.ai will forever change the music intelligence industry, as it allows scientists to execute powerful analytics and machine learning algorithms at scale on a huge data-set without the need to write a single-line of code."
"No more time and money wasted on finding, cleansing, and storing this hard-to-find music data, and especially, on the engineering effort needed to operate these complex graph computing infrastructure - which require deep domain expertise and PhD-level talent."
GigaOm has more on the product
Developers can find out more here.