| Titre : | Recommender system development and User Data analytics for DataGalaxy platform | | Type de document : | projet fin études | | Auteurs : | Maryeme BAYRI, Auteur | | Langues : | Français (fre) | | Catégories : | e-Management et Business Intelligence
| | Mots-clés : | DataGalaxy, recommendation, logs analysis, SQL, ElasticSearch, PowerBI,
monitor | | Index. dĂ©cimale : | 2181/19 | | RĂ©sumĂ© : | Recommender systems play an important role in enhancing users’ experience. It allows them
to easily use and navigate platforms and applications in the large sea of content. Many users
have a hard time not only with finding something they want but even with figuring out what is
it that they want in the first place and what are their choices. And even when they are provided
with aiding tools like RS, they still find themselves lost and confused. The only way for a
company to get that knowledge and understand the needs of their users is through analyzing
their behavior and activities.
This work consists of integrating a recommendation engine into DataGalaxy’s platform that
aims to aid users in the association between their data and their business definitions and
descriptions as well as analyzing user navigation logs in order to grasp their activities, evaluate
the usage rate and satisfaction of the developed RS and base on that analysis in the decision
making process and in defining future projects.
Our RS maps between the elements to associate and provides suggestions for those with the
highest score of similarity by inserting them in an ElasticSearch index whenever the user adds
or modifies a business definition or data in the platform. Then, by retrieving user navigation
logs stored in a SQL database and creating reports and a dashboard on PowerBI, we are able to
monitor the developed RS as well as other features and track their performance.
Through our analysis, we found that the RS has a high rate of accepted suggestions compared
to the declination rate and also highlighted some clients’ need for more guidance and assistance
in using the solution.
In this work, we first present all the steps required to ensure the management of our project.
Then, we report on the details of our approach and the results we obtained using log
analysis.
|
Recommender system development and User Data analytics for DataGalaxy platform [projet fin études] / Maryeme BAYRI, Auteur . - [s.d.]. Langues : Français ( fre) | Catégories : | e-Management et Business Intelligence
| | Mots-clés : | DataGalaxy, recommendation, logs analysis, SQL, ElasticSearch, PowerBI,
monitor | | Index. dĂ©cimale : | 2181/19 | | RĂ©sumĂ© : | Recommender systems play an important role in enhancing users’ experience. It allows them
to easily use and navigate platforms and applications in the large sea of content. Many users
have a hard time not only with finding something they want but even with figuring out what is
it that they want in the first place and what are their choices. And even when they are provided
with aiding tools like RS, they still find themselves lost and confused. The only way for a
company to get that knowledge and understand the needs of their users is through analyzing
their behavior and activities.
This work consists of integrating a recommendation engine into DataGalaxy’s platform that
aims to aid users in the association between their data and their business definitions and
descriptions as well as analyzing user navigation logs in order to grasp their activities, evaluate
the usage rate and satisfaction of the developed RS and base on that analysis in the decision
making process and in defining future projects.
Our RS maps between the elements to associate and provides suggestions for those with the
highest score of similarity by inserting them in an ElasticSearch index whenever the user adds
or modifies a business definition or data in the platform. Then, by retrieving user navigation
logs stored in a SQL database and creating reports and a dashboard on PowerBI, we are able to
monitor the developed RS as well as other features and track their performance.
Through our analysis, we found that the RS has a high rate of accepted suggestions compared
to the declination rate and also highlighted some clients’ need for more guidance and assistance
in using the solution.
In this work, we first present all the steps required to ensure the management of our project.
Then, we report on the details of our approach and the results we obtained using log
analysis.
|
|
Faire une suggestion Affiner la recherche
Recommender system development and User Data analytics for DataGalaxy platform / Maryeme BAYRI
