M. Guèye

A parameter-free algorithm for an optimized tag recommendation list size

Tag recommendation is a major aspect of collaborative tagging systems. It aims to recommend suitable tags to a user for tagging an item. One of its main challenges is the effectiveness of its recommendations. Existing works focus on techniques for retrieving the most relevant tags to give beforehand, with a fixed number of tags in each recommended list. In this paper, we try to optimize the number of recommended tags in order to improve the efficiency of the recommendations. We propose a parameter-free algorithm for determining the optimal size of the recommended list. Thus we introduced some relevance measures to find the most relevant sublist from a given list of recommended tags. More precisely, we improve the quality of our recommendations by discarding some unsuitable tags and thus adjusting the list size. Our solution is an add-on one, which can be implemented on top of many kinds of tag recommenders. The experiments we did on five datasets, using four categories of tag recommenders, demonstrate the efficiency of our technique. For instance, the algorithm we propose outperforms the results of the task 2 of the ECML PKDD Discovery Challenge 20091. By using the same tag recommender than the winners of the contest, we reach a F1 measure of 0.366 while the latter got 0.356. Thus, our solution yields significant improvements on the lists obtained from the tag recommenders.

Database replication in large scale systems: optimizing the number of replicas

In distributed systems, replication is used for ensuring availability and increasing performances. However, the heavy workload of distributed systems such as web2.0 applications or Global Distribution Systems, limits the benefit of replication if its degree (i.e., the number of replicas) is not controlled. Since every replica must perform all updates eventually, there is a point beyond which adding more replicas does not increase the throughput, because every replica is saturated by applying updates. Moreover, if the replication degree exceeds the optimal threshold, the useless replica would generate an overhead due to extra communication messages. In this paper, we propose a suitable replication management solution in order to reduce useless replicas. To this end, we define two mathematical models which approximate the appropriate number of replicas to achieve a given level of performance. Moreover, we demonstrate the feasibility of our replication management model through simulation. The results expose the effectiveness of our models and their accuracy.

Contribution of Dehydration and Malnutrition to the Mortality of Children 0-59 Month of Age in a Senegalese Pediatric Hospital

In-hospital mortality is an indicator of the quality of care. We analyzed the mortality of under five years children of Pediatric ward of Aristide Le Dantec teaching hospital to update our data, after an previous study conducted ten years earlier. Methods : This was a retrospective study involving children 0-59 months of age, hospitalized from January 1, 2012 to December 31, 2012. For each child, nutritional status was assessed according to 2006 World Health Organization growth standards; clinical and biological data were recorded. The outcome of the disease was specified. Bivariate and multivariable were used to identify risk factors for death. Results : 393 children were included. Overall mortality rate was 10% (39/393). Factors associated with death were severe wasting [OR = 8.27, 95% CI [3.79-18], male gender (OR = 2.98, 95% CI [1.25-7.1]), dehydration (OR = 5.4, 95% CI [2.54-13.43]) in the model using the weight-for- height z score, male gender (OR = 2.5, 95% CI [1.11-5.63]), dehydration (OR = 8.43, 95% CI [3.83-18.5]) in using the height- for- age z score, male gender (OR = 2.7, 95% CI [1.19-6.24]), dehydration (OR = 7.5, 95% CI [3.39-16.76]), severe underweight (OR = 2.4, 95% CI [1.11-5.63]), in the model using the weight-for- age z score, and male gender (OR = 2.5, 95% CI [1.11-5.63]), dehydration (OR = 8.43, 95% CI [3.83-18.5]) in that using MUAC. Dehydration and malnutrition are two independent factors of mortality. Our management protocols of dehydration and malnutrition have to be updated. Screening malnutrition has to be done systematically for each child by anthropometric measurements using WHO growth standards.

Complication rare d’une méningite néonatale à Streptocoque B : le diabète insipide central

Le diabète insipide central par déficit en hormone antidiurétique est une maladie rare de la période néonatale, souvent secondaire à une affection neurologique sévère [1,2] parfois d’origine génétique [3]. L’efficacité du traitement substitutif par desmopressine en période néonatale a été prouvée [4]. Nous présentons ici le cas d’un nouveau-né ayant développé un diabète insipide central au décours d’une méningite à Streptocoque B. Il s’agissait d’un garçon, second enfant d’une fratrie de deux enfants, né eutrophique à terme après une grossesse et un accouchement sans particularités. Il était admis à 15 j de vie pour un syndrome méningé avec des convulsions tonicocloniques généralisées et une détresse respiratoire sévère. C’était associé à un syndrome infectieux et avait justifié une antibiothérapie empirique en ambulatoire 48 h avant l’hospitalisation. La ponction lombaire (PL) confirmait l’existence d’une méningite suppurée en ramenant un liquide céphalorachidien (LCR) louche avec une pléı̈ocytose à 1820 leucocytes par mm3 avec prédominance de polynucléaires neutrophiles (PNN) altérés (66 %), une hyperalbuminorachie à 1,09 g/L et une hypoglycorachie à 0,16 g/l. L’examen au latex mettait en évidence les antigènes du Streptocoque B mais la culture du LCR était négative (méningite décapitée). L’hémogramme mettait en évidence une leuconeutropénie à 3820 par mm3. L’ionogramme sanguin, en particulier la