A obra é uma coletânea dos melhores trabalhos desenvolvidos no final dos cursos de especialização em Gestão da Cadeia de Suprimento e Logística do LALT / FEC / Unicamp. Volume 1 da série LALT de Produção Aplicada
A obra é uma coletânea dos melhores trabalhos desenvolvidos no final dos cursos de especialização em Gestão da Cadeia de Suprimento e Logística do LALT / FEC / Unicamp. Volume 1 da série LALT de Produção Aplicada
Lucene/Solr Revolution 2015: Where Search Meets Machine LearningJoaquin Delgado PhD.
Search engines have focused on solving the document retrieval problem, so their scoring functions do not handle naturally non-traditional IR data types, such as numerical or categorical. Therefore, on domains beyond traditional search, scores representing strengths of associations or matches may vary widely. As such, the original model doesn’t suffice, so relevance ranking is performed as a two-phase approach with 1) regular search 2) external model to re-rank the filtered items. Metrics such as click-through and conversion rates are associated with the users’ response to items served. The predicted selection rates that arise in real-time can be critical for optimal matching. For example, in recommender systems, predicted performance of a recommended item in a given context, also called response prediction, is often used in determining a set of recommendations to serve in relation to a given serving opportunity. Similar techniques are used in the advertising domain. To address this issue the authors have created ML-Scoring, an open source framework that tightly integrates machine learning models into a popular search engine (SOLR/Elasticsearch), replacing the default IR-based ranking function. A custom model is trained through either Weka or Spark and it is loaded as a plugin used at query time to compute custom scores.
2. Yes, I think my technical skills have gotten better because have had more practice will
the programs we had to use and had more time to work out how to use the different
functions that I needed for different elements in my magazine. They have got better
because we’ve had more time to be taught to use the functions. I definitely thought
about my content more in my music magazine, mainly because there was more
contents to think about because we had to do more pages. However, I had to think
about the content more because I wanted the magazine to fit as a whole better then
my college magazine did so I had to make sure all the elements of the magazine
worked together.
3. • I definitely think I was more organised, because I managed to finish before the
deadline and make 2 extra pages so my time management was very good, I didn’t
lose any of my file because I organised them in to the correct folders.
•
I was a lot more ambitious because the college magazine was quite simple and
boring, whereas I put more thought in the colours and images. Also, I think I was quite
ambitious with the content of the articles.
• I think I was more creative with my music magazine then I was with my college one
because of the colours, images.
• I didn’t use any of the ideas from my college magazine because looking back on it, I
didn’t like my college magazine very much.