Using visual and text features for direct marketing on multimedia messaging services domain

Traditionally, direct marketing companies have relied on pre-testing to select the best offers to send to their audience. Companies systematically dispatch the offers under consideration to a limited sample of potential buyers, rank them with respect to their performance and, based on this ranking, decide which offers to send to the wider population. Though this pre-testing process is simple and widely used, recently the industry has been under increased pressure to further optimize learning, in particular when facing severe time and learning space constraints. The main contribution of the present work is to demonstrate that direct marketing firms can exploit the information on visual content to optimize the learning phase. This paper proposes a two-phase learning strategy based on a cascade of regression methods that takes advantage of the visual and text features to improve and accelerate the learning process. Experiments in the domain of a commercial Multimedia Messaging Service (MMS) show the effectiveness of the proposed methods and a significant improvement over traditional learning techniques. The proposed approach can be used in any multimedia direct marketing domain in which offers comprise both a visual and text component.

Degree-Optimal Routing for P2P Systems

We define a family of Distributed Hash Table systems whose aim is to combine the routing efficiency of randomized networks—e.g. optimal average path length O(log ² n/δlog δ) with δ degree—with the programmability and startup efficiency of a uniform overlay—that is, a deterministic system in which the overlay network is transitive and greedy routing is optimal. It is known that Ω(log n) is a lower bound on the average path length for uniform overlays with O(log n) degree (Xu et al., IEEE J. Sel. Areas Commun. 22(1), 151–163, 2004). Our work is inspired by neighbor-of-neighbor (NoN) routing, a recently introduced variation of greedy routing that allows us to achieve optimal average path length in randomized networks. The advantage of our proposal is that of allowing the NoN technique to be implemented without adding any overhead to the corresponding deterministic network. We propose a family of networks parameterized with a positive integer c which measures the amount of randomness that is used. By varying the value c, the system goes from the deterministic case (c=1) to an “almost uniform” system. Increasing c to relatively low values allows for routing with asymptotically optimal average path length while retaining most of the advantages of a uniform system, such as easy programmability and quick bootstrap of the nodes entering the system. We also provide a matching lower bound for the average path length of the routing schemes for any c. This work was partially supported by the Italian FIRB project “WEB-MINDS” (Wide-scalE, Broadband MIddleware for Network Distributed Services), .     

An empirical evaluation of active learning strategies for profile elicitation in a conversational recommender system

Journal of Intelligent Information Systems - Conversational Recommender Systems have received widespread attention in both research and practice. They assist people in finding relevant and...     

Prototyping and modeling of the centrifugal casting process for paraffin waxes

Centrifugal casting is a technology used for manufacturing hybrid rocket paraffin grains. This technology helps avoiding voids formation inside the solid paraffin as it cools. Voids are formed because of air bubbles being entrapped while pouring and because the liquid wax shrinks by 17–19% upon cooling. In this work, the centrifugal casting process for the manufacturing of paraffin cylinders was prototyped at two different scales considering critical casting issues. The effects of process parameters (rotational speed, melt temperature, and flow rate) on the tensile properties of the manufactured grains were analyzed. The results of the optimization conducted at the lower scale (2.5?kg) were up scaled to manufacture 25?kg grains. The resulting mechanical properties complied with the design specifications, and they were better than those characterized from the gravity cast wax. A numerical model of growth and dissolution of bubbles during the process was then developed to predict the quality of the castings. The numerical results showed how increasing the mold rotational speed up to 1800?rpm reduced the removal time. However, compared to grains solidification time, the predicted removal times were much shorter, proving the advantage of centrifugal casting in counteracting voids formation.     

Monitoring the Ratio of Two Normal Variables Using EWMA Type Control Charts

In many fields, there is the need to monitor quality characteristics defined as the ratio of two random variables. The design and implementation of control charts directly monitoring the ratio stability is required for the continuous surveillance of these quality characteristics. In this paper, we propose two one‐sided exponentially weighted moving average (EWMA) charts with subgroups having sample size n > 1 to monitor the ratio of two normal random variables. The optimal EWMA smoothing constants, control limits, and ARLs have been computed for different values of the in‐control ratio and correlation between the variables and are shown in several figures and tables to discuss the statistical performance of the proposed one‐sided EWMA charts. Both deterministic and random shift sizes have been considered to test the two one‐sided EWMA charts' sensitivity. The obtained results show that the proposed one‐sided EWMA control charts are more sensitive to process shifts than other charts already proposed in the literature. The practical application of the proposed control schemes is discussed with an illustrative example. Copyright © 2015 John Wiley & Sons, Ltd.     

Nanostructured Materials for Room‐Temperature Gas Sensors

Sensor technology has an important effect on many aspects in our society, and has gained much progress, propelled by the development of nanoscience and nanotechnology. Current research efforts are directed toward developing high‐performance gas sensors with low operating temperature at low fabrication costs. A gas sensor working at room temperature is very appealing as it provides very low power consumption and does not require a heater for high‐temperature operation, and hence simplifies the fabrication of sensor devices and reduces the operating cost. Nanostructured materials are at the core of the development of any room‐temperature sensing platform. The most important advances with regard to fundamental research, sensing mechanisms, and application of nanostructured materials for room‐temperature conductometric sensor devices are reviewed here. Particular emphasis is given to the relation between the nanostructure and sensor properties in an attempt to address structure–property correlations. Finally, some future research perspectives and new challenges that the field of room‐temperature sensors will have to address are also discussed.     

Dynamics and self-assembly of bio-functionalized gold nanoparticles in solution: Reactive molecular dynamics simulations

The self-assembling properties, stability, and dynamics of hybrid nanocarriers (gold nanoparticles (AuNPs) functionalized with cysteine-based peptides) in solution are studied through a series of classical molecular dynamics simulations based on a recently parametrized reactive force field. The results reveal, at the atomic level, all the details regarding the peptide adsorption mechanisms, nanoparticle stabilization, aggregation, and sintering. The data confirm and explain the experimental findings and disclose aspects that cannot be scrutinized by experiments. The biomolecules are both chemisorbed and physisorbed; self-interactions of the adsorbates and formation of stable networks of interconnected molecules on the AuNP surfaces limit substrate reconstructions, protect the AuNPs from the action of the solvent, and prevent direct interactions of the gold surfaces. The possibility of agglomeration of the functionalized nanoparticles, compared with the sintering of the bare supports in a water solution, is demonstrated through relatively long simulations and fast steered dynamics. The analysis of the trajectories reveals that the AuNPs were well stabilized by the peptides. This prevented particle sintering and kept the particles far apart; however, part of their chains could form interconnections (crosslinks) between neighboring gold vehicles. The excellent agreement of these results with the literature confirm the reliability of the method and its potential application to the modeling of more complex materials relevant to the biomedical sector.

Open image in new window

     

Logic based methods for SNPs tagging and reconstruction

SNPs are positions of the DNA sequences where the differences among individuals are embedded. The knowledge of such SNPs is crucial for disease association studies, but even if the number of such positions is low (about 1% of the entire sequence), the cost to extract the complete information is actually very high. Recent studies have shown that DNA sequences are structured into blocks of positions, that are conserved during evolution, where there is strong correlation among values (alleles) of different loci. To reduce the cost of extracting SNPs information, the block structure of the DNA has suggested to limit the process to a subset of SNPs, the so-called Tag SNPs, that are able to maintain the most of the information contained in the whole sequence. In this paper, we apply a technique for feature selection based on integer programming to the problem of Tag SNP selection. Moreover, to test the quality of our approach, we consider also the problem of SNPs reconstruction, i.e. the problem of deriving unknown SNPs from the value of Tag SNPs and propose two reconstruction methods, one based on a majority vote and the other on a machine learning approach. We test our algorithm on two public data sets of different nature, providing results that are, when comparable, in line with the related literature. One of the interesting aspects of the proposed method is to be found in its capability to deal simultaneously with very large SNPs sets, and, in addition, to provide highly informative reconstruction rules in the form of logic formulas.     

Blind maximum likelihood estimation of traffic matrices under long-range dependent traffic

A new method, based on the maximum likelihood principle, through the numerical Expectation–Maximization algorithm, is proposed to estimate traffic matrices when traffic exhibits long-range dependence. The methods proposed so far in the literature do not account for long-range dependence. The method proposed in the present paper also provides an estimate of the Hurst parameter. Simulation results show that: (i) the estimate of the traffic matrix is more efficient than those obtained via existing techniques; (ii) the estimation error of the traffic matrix is lower for larger values of the true traffic intensity; (iii) the estimate of the Hurst parameter is slightly negatively biased.