Interest-based recommendations for business intelligence users

Drushku, Krista; Aligon, Julien; Labroche, Nicolas; Marcel, Patrick; Peralta, Verónika

doi:10.1016/j.is.2018.08.004

Cited by 16 publications

(15 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Learning distances between actions corresponds to a traditional metric learning problem. Classical approaches addressing this problem include (i) transformation to classification task [5,9,14], or (ii) weekly-supervised metric learning [15]. Regarding (i), distances are usually linear combinations of feature-wise distances between objects, obtained by fitting a linear classifier (e.g., SVM) over pairs of objects labeled positively if objects should be in the same group and negatively otherwise.…”

Section: Computing Action Profiles (P 1 )mentioning

confidence: 99%

“…In what follows, we denote by ∼ a similarity function between actions. We consider that there is a match between a recommended action and an expected action from the pathway if their similarity exceeds a threshold that is a parameter of the precision and recall measures, similar to what is done in [1,5].…”

Section: Effectiveness Of Prsmentioning

confidence: 99%

“…Our way to learn action profiles is similar to user intent discovery [5,9,14]. For instance, Guha et al [9] discover user intent behind web searches, and obtain content relevant to users' long-term interests.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

A Chain Composite Item Recommender for Lifelong Pathways

Chanson

Devogèle

Labroche

et al. 2021

Big Data Analytics and Knowledge Discovery

Self Cite

View full text Add to dashboard Cite

This work addresses the problem of recommending lifelong pathways, i.e., sequences of actions pertaining to health, social or professional aspects, for fulfilling a personal lifelong project. This problem raises some specific challenges, since the recommendation process is constrained by the user profile, the time they can devote to the actions in the pathway, the obligation to smooth the learning curve of the user. We model lifelong pathways as particular chain composite items and formalize the recommendation problem as a form of orienteering problem. We adapt classical evaluation criteria for measuring the quality of the recommended pathways. We experiment with both artificial and real datasets, showing our approach is a promising building block of an interactive lifelong pathways recommender system.

show abstract

Section: Computing Action Profiles (P 1 )mentioning

confidence: 99%

Section: Effectiveness Of Prsmentioning

confidence: 99%

See 1 more Smart Citation

A Chain Composite Item Recommender for Lifelong Pathways

Chanson

Devogèle

Labroche

et al. 2021

Big Data Analytics and Knowledge Discovery

Self Cite

View full text Add to dashboard Cite

show abstract

“…Assume now that the Ministry of Labor, based on the data of a previous census, has set-up some goals for the improvement of labor. Assume also that per combination of education type and workplace, a specific goal, say W eekly W orking T arget, has been assigned, saying that if the average number of weekly hours at work is in the area of [40][41][42][43][44][45][46][47][48][49][50][51][52][53][54][55] for any category, then this is Expected behavior, whereas any other amount outside this domain is either Low, or Excessive. This is exactly what business analysts call a Key Performance Indicator (KPI) Then, assume that the analyst of the Ministry wishes to evaluate the situation based on these goals, and in fact, in more details than the aggregate summary of Example 1.…”

Section: The Vision In a Nutshellmentioning

confidence: 99%

“…The authors of [47,31] use the query log of previous users to find similar queries which can give information to user who may not know it is available. Drushku et al [49] use the query log of previous users to detect user intents and recommend queries that fit the user's current intents.…”

Section: Query Recommendationsmentioning

confidence: 99%

Beyond roll-up’s and drill-down’s: An intentional analytics model to reinvent OLAP

Vassiliadis

Marcel

Rizzi

2019

Information Systems

Self Cite

View full text Add to dashboard Cite

This paper structures a novel vision for OLAP by fundamentally redefining several of the pillars on which OLAP has been based for the last 20 years. We redefine OLAP queries, in order to move to higher degrees of abstraction from roll-up's and drill-down's, and we propose a set of novel intentional OLAP operators, namely, describe, assess, explain, predict, and suggest, which express the user's need for results. We fundamentally redefine what a query answer is, and escape from the constraint that the answer is a set of tuples; on the contrary, we complement the set of tuples with models (typically, but not exclusively, results of data mining algorithms over the involved data) that concisely represent the internal structure or correlations of the data. Due to the diverse nature of the involved models, we come up (for the first time ever, to the best of our knowledge) with a unifying framework for them, that places its pillars on the extension of each data cell of a cube with information about the models that pertain to it -practically converting the small parts that build up the models to data that annotate each cell. We exploit this data-to-model mapping to provide highlights of the data, by isolating data and models that maximize the delivery of new information to the user. We introduce a novel method for assessing the surprise that a new query result brings to the user, with respect to the information contained in previous results the user has seen via a new interestingness measure. The individual parts of our proposal are integrated in a new data model for OLAP, which we call the Intentional Analytics Model. We complement our contribution with a list of significant open problems for the community to address.

show abstract