'On the Effects of Idiotypic Interactions for Recommendation Communities in Artificial Immune Systems'

It has previously been shown that a recommender based on immune system idiotypic principles can outperform one based on correlation alone. This paper reports the results of work in progress, where we undertake some investigations into the nature of this beneficial effect. The initial findings are that the immune system recommender tends to produce different neighbourhoods, and that the superior performance of this recommender is due partly to the different neighbourhoods, and partly to the way that the idiotypic effect is used to weight each neighbour's recommendations.

On The Effects of Idiotypic Interactions for Recommendation Communities in Artificial Immune Systems

It has previously been shown that a recommender based on immune system idiotypic principles can outperform one based on correlation alone. This paper reports the results of work in progress, where we undertake some investigations into the nature of this beneficial effect. The initial findings are that the immune system recommender tends to produce different neighbourhoods, and that the superior performance of this recommender is due partly to the different neighbourhoods, and partly to the way that the idiotypic effect is used to weight each neighbour’s recommendations.

Dating of sediments in the Biscay bay: Implication for pollution chronology

Important historical informations on the temporal changes of anthropogenic pollution in marine environment can be assessed using sediment analysis. Dating is a crucial prerequisite to reconstruct pollution events, to calculate fluxes, and thus to allow comparison between different sites. This work presents estimates of accumulation rates of sediments in the Bay of Biscay. Fives cores were collected during RIKEAU 2002 cruise on board o/v Thalia in order to study temporal changes in PAH and organohalogens compounds content of sediment. We compare chronostratigraphic estimates on cores derived from the natural radionuclide 210Pb in excess with estimates from the known times of introduction of the artificial radionuclide 137Cs to the environment. 210Pb, 226Ra and 137Cs were measured directly by non-destructive gamma spectrometry using a well type γ-detector. Total 210Pb and 226Ra activities vary from 30 to 150 mBq g-1, and 20 to 36 mBq g-1 respectively; 137Cs presents lower levels (< 5 mBq g-1). Profiles of 210Pb in three cores present a well mixed layer, from 2-3 to 10 cm, in the uppermost sediments, followed by an exponential decrease of activities, suitable for the determination of sedimentation rates. Under constant flux and sedimentation rate assumptions, vertical accretion rates derived from 210Pb present a large range from nearly 0.1 cm yr-1 up to almost 0.3 cm yr-1. Differences are mainly due to relative position of studied cores regarding the muddy patch. Although the moderate level of 137Cs limits the accuracy of this dating method, profiles of 137Cs with depth strengthen mean rates derived from 210Pb data. The implication of this dating on pollutant inputs in sediments of the Bay of Biscay is briefly discussed.

The role of context in image annotation and recommendation

With the rise of smart phones, lifelogging devices (e.g. Google Glass) and popularity of image sharing websites (e.g. Flickr), users are capturing and sharing every aspect of their life online producing a wealth of visual content. Of these uploaded images, the majority are poorly annotated or exist in complete semantic isolation making the process of building retrieval systems difficult as one must firstly understand the meaning of an image in order to retrieve it. To alleviate this problem, many image sharing websites offer manual annotation tools which allow the user to “tag” their photos, however, these techniques are laborious and as a result have been poorly adopted; Sigurbjörnsson and van Zwol (2008) showed that 64% of images uploaded to Flickr are annotated with < 4 tags. Due to this, an entire body of research has focused on the automatic annotation of images (Hanbury, 2008; Smeulders et al., 2000; Zhang et al., 2012a) where one attempts to bridge the semantic gap between an image’s appearance and meaning e.g. the objects present. Despite two decades of research the semantic gap still largely exists and as a result automatic annotation models often offer unsatisfactory performance for industrial implementation. Further, these techniques can only annotate what they see, thus ignoring the “bigger picture” surrounding an image (e.g. its location, the event, the people present etc). Much work has therefore focused on building photo tag recommendation (PTR) methods which aid the user in the annotation process by suggesting tags related to those already present. These works have mainly focused on computing relationships between tags based on historical images e.g. that NY and timessquare co-exist in many images and are therefore highly correlated. However, tags are inherently noisy, sparse and ill-defined often resulting in poor PTR accuracy e.g. does NY refer to New York or New Year? This thesis proposes the exploitation of an image’s context which, unlike textual evidences, is always present, in order to alleviate this ambiguity in the tag recommendation process. Specifically we exploit the “what, who, where, when and how” of the image capture process in order to complement textual evidences in various photo tag recommendation and retrieval scenarios. In part II, we combine text, content-based (e.g. # of faces present) and contextual (e.g. day-of-the-week taken) signals for tag recommendation purposes, achieving up to a 75% improvement to precision@5 in comparison to a text-only TF-IDF baseline. We then consider external knowledge sources (i.e. Wikipedia & Twitter) as an alternative to (slower moving) Flickr in order to build recommendation models on, showing that similar accuracy could be achieved on these faster moving, yet entirely textual, datasets. In part II, we also highlight the merits of diversifying tag recommendation lists before discussing at length various problems with existing automatic image annotation and photo tag recommendation evaluation collections. In part III, we propose three new image retrieval scenarios, namely “visual event summarisation”, “image popularity prediction” and “lifelog summarisation”. In the first scenario, we attempt to produce a rank of relevant and diverse images for various news events by (i) removing irrelevant images such memes and visual duplicates (ii) before semantically clustering images based on the tweets in which they were originally posted. Using this approach, we were able to achieve over 50% precision for images in the top 5 ranks. In the second retrieval scenario, we show that by combining contextual and content-based features from images, we are able to predict if it will become “popular” (or not) with 74% accuracy, using an SVM classifier. Finally, in chapter 9 we employ blur detection and perceptual-hash clustering in order to remove noisy images from lifelogs, before combining visual and geo-temporal signals in order to capture a user’s “key moments” within their day. We believe that the results of this thesis show an important step towards building effective image retrieval models when there lacks sufficient textual content (i.e. a cold start).

Factors influencing the recommendation of the Human Papillomavirus vaccine by South African doctors working in a tertiary hospital.

Background: In South Africa, HPV vaccination programme has been incorporated recently in the school health system. Since doctors are the most trusted people regarding health issues in general, their knowledge and attitudes regarding HPV infections and vaccination are very important for HPV vaccine program nationally. Objective: The objective of this study was to investigate factors contributing to recommendation of HPV vaccines to the patients. Methods: This was a quantitative cross-sectional study conducted among 320 doctors, using a self-administered anonymous questionnaire. Results: All the doctors were aware of HPV and knew that HPV is transmitted sexually. Their overall level of knowledge regarding HPV infections and HPV vaccine was poor. But the majority intended to prescribe the vaccine to their patients. It was found that doctors who knew that HPV 6 and 11 are responsible for >90% of anogenital warts, their patients would comply with the counselling regarding HPV vaccination, and received sufficient information about HPV vaccination were 5.68, 4.91 and 4.46 times respectively more likely to recommend HPV vaccination to their patients, compared to their counterparts (p<0.05). Conclusion: There was a knowledge gap regarding HPV infection and HPV vaccine among the doctors.

SCOR WG 142: Quality Control Procedures for Oxygen and Other Biogeochemical Sensors on Floats and Gliders. Recommendation for oxygen measurements from Argo floats, implementation of in-air-measurement routine to assure highest long-term accuracy

Recommendation for Oxygen Measurements from Argo Floats: Implementation of In-Air-Measurement Routine to Assure Highest Long-term Accuracy As Argo has entered its second decade and chemical/biological sensor technology is improving constantly, the marine biogeochemistry community is starting to embrace the successful Argo float program. An augmentation of the global float observatory, however, has to follow rather stringent constraints regarding sensor characteristics as well as data processing and quality control routines. Owing to the fairly advanced state of oxygen sensor technology and the high scientific value of oceanic oxygen measurements (Gruber et al., 2010), an expansion of the Argo core mission to routine oxygen measurements is perhaps the most mature and promising candidate (Freeland et al., 2010). In this context, SCOR Working Group 142 “Quality Control Procedures for Oxygen and Other Biogeochemical Sensors on Floats and Gliders” (www.scor-int.org/SCOR_WGs_WG142.htm) set out in 2014 to assess the current status of biogeochemical sensor technology with particular emphasis on float-readiness, develop pre- and post-deployment quality control metrics and procedures for oxygen sensors, and to disseminate procedures widely to ensure rapid adoption in the community.

ESR dating of pleistocene mammal teeth and its implications for the biostratigraphy and geological evolution of the coastal plain, Rio Grande do Sul, southern Brazil

The fossiliferous deposits in the coastal plain of the Rio Grande do Sul State, Southern Brazil, have been known since the late XIX century; however, the biostratigraphic and chronostratigraphic context is still poorly understood. The present work describes the results of electron spin resonance (ESR) dating in eleven fossil teeth of three extinct taxa (Toxodon platensis, Stegomastodon waringi and Hippidion principale) collected along Chui Creek and nearshore continental shelf, in an attempt to assess more accurately the ages of the fossils and its deposits. This method is based upon the analysis of paramagnetic defects found in biominerals, produced by ionizing radiation emitted by radioactive elements present in the surrounding sediment and by cosmic rays. Three fossils from Chui Creek, collected from the same stratigraphic horizon, exhibit ages between (42 +/- 3) Ka and (34 +/- 7) Ka, using the Combination Uptake model for radioisotopes uptake, while a incisor of Toxodon platensis collected from a stratigraphic level below is much older. Fossils from the shelf have ages ranging from (7 +/- 1) 10(5) Ka to (18 +/- 3) Ka, indicating the mixing of fossils of different epochs. The origin of the submarine fossiliferous deposits seems to be the result of multiple reworking and redeposition cycles by sea-level changes caused by the glacial-interglacial cycles during the Quaternary. The ages indicate that the fossiliferous outcrops at Chui Creek are much younger than previously thought, and that the fossiliferous deposits from the continental shelf encompass Ensenadan to late Lujanian ages (middle to late Pleistocene). (C) 2009 Elsevier Ltd and INQUA. All rights reserved.

Characterization and age determination of Quaternary volcanism in the southern Ankaratra region (central Madagascar) through novel approaches in luminescence dating

This work introduces two novel approaches for the application of luminescence dating techniques to Quaternary volcanic eruptions: crystalline xenoliths from lava flows are demonstrated to be basically suitable for luminescence dating, and a set of phreatic explosion deposits from the Late Quaternary Vakinankaratra volcanic field in central Madagascar is successfully dated with infrared stimulated luminescence (IRSL). Using a numerical model approach and experimental verification, the potential for thermal resetting of luminescence signals of xenoliths in lava flows is demonstrated. As microdosimetry is an important aspect when using sample material extracted from crystalline whole rocks, autoradiography using image plates is introduced to the field of luminescence dating as a method for detection and assessment of spatially resolved radiation inhomogeneities. Determinations of fading rates of feldspar samples have been observed to result in aberrant g-values if the pause between preheat and measurement in the delayed measurements was kept short. A systematic investigation reveals that the phenomenon is caused by the presence of three signal components with differing individual fading behaviour. As this is restricted to short pauses, it is possible to determine a minimal required delay between preheating and measurement after which the aberrant behaviour disappears. This is applied in the measuring of 12 samples from phreatic explosion deposits from the Antsirabe – Betafo region in the Late Quaternary Vakinankaratra volcanic field. The samples were taken from stratigraphically correlatable sections and appear to represent at least three phreatic events, one of which created the Lac Andraikiba maar near Antsirabe. The obtained ages indicate that the eruptive activity in the region started in the Late Pleistocene between 113.9 and 99.6 ka. A second layer in the Betafo area is dated at approximately 73 ka and the Lac Andraikiba deposits give an age between 63.9 and 50.7 ka. The youngest phreatic layer is dated between 33.7 and 20.7 ka. These ages are the first recorded direct ages of such volcanic deposits, as well as the first and only direct ages for the Late Quaternary volcanism in the Vakinankaratra volcanic field. This illustrates the huge potential of this new method for volcanology and geochronology, as it enables direct numerical dating of a type of volcanic deposit which has not been successfully directly dated by any other method so far.

A COMPARATIVE STUDY AND EVALUATION OF COLLABORATIVE RECOMMENDATION SYSTEMS

Consumers currently enjoy a surplus of goods (books, videos, music, or other items) available to purchase. While this surplus often allows a consumer to find a product tailored to their preferences or needs, the volume of items available may require considerable time or effort on the part of the user to find the most relevant item. Recommendation systems have become a common part of many online business that supply users books, videos, music, or other items to consumers. These systems attempt to provide assistance to consumers in finding the items that fit their preferences. This report presents an overview of recommendation systems. We will also briefly explore the history of recommendation systems and the large boost that was given to research in this field due to the Netflix Challenge. The classical methods for collaborative recommendation systems are reviewed and implemented, and an examination is performed contrasting the complexity and performance among the various models. Finally, current challenges and approaches are discussed.

Collaborative filtering and deep learning based recommendation system for cold start items

Recommender system is a specific type of intelligent systems, which exploits historical user ratings on items and/or auxiliary information to make recommendations on items to the users. It plays a critical role in a wide range of online shopping, e-commercial services and social networking applications. Collaborative filtering (CF) is the most popular approaches used for recommender systems, but it suffers from complete cold start (CCS) problem where no rating record are available and incomplete cold start (ICS) problem where only a small number of rating records are available for some new items or users in the system. In this paper, we propose two recommendation models to solve the CCS and ICS problems for new items, which are based on a framework of tightly coupled CF approach and deep learning neural network. A specific deep neural network SADE is used to extract the content features of the items. The state of the art CF model, timeSVD++, which models and utilizes temporal dynamics of user preferences and item features, is modified to take the content features into prediction of ratings for cold start items. Extensive experiments on a large Netflix rating dataset of movies are performed, which show that our proposed recommendation models largely outperform the baseline models for rating prediction of cold start items. The two proposed recommendation models are also evaluated and compared on ICS items, and a flexible scheme of model retraining and switching is proposed to deal with the transition of items from cold start to non-cold start status. The experiment results on Netflix movie recommendation show the tight coupling of CF approach and deep learning neural network is feasible and very effective for cold start item recommendation. The design is general and can be applied to many other recommender systems for online shopping and social networking applications. The solution of cold start item problem can largely improve user experience and trust of recommender systems, and effectively promote cold start items.

Collaborative filtering and deep learning based hybrid recommendation for cold start problem

Recommender systems (RS) are used by many social networking applications and online e-commercial services. Collaborative filtering (CF) is one of the most popular approaches used for RS. However traditional CF approach suffers from sparsity and cold start problems. In this paper, we propose a hybrid recommendation model to address the cold start problem, which explores the item content features learned from a deep learning neural network and applies them to the timeSVD++ CF model. Extensive experiments are run on a large Netflix rating dataset for movies. Experiment results show that the proposed hybrid recommendation model provides a good prediction for cold start items, and performs better than four existing recommendation models for rating of non-cold start items.

Improved recommendation of photo-taking locations using virtual ratings

We consider the task of collaborative recommendation of photo-taking locations. We use datasets of geotagged photos. We map their locations to a location grid using a geohashing algorithm, resulting in a user x location implicit feedback matrix. Our improvements relative to previous work are twofold. First, we create virtual ratings by spreading users' preferences to neighbouring grid locations. This makes the assumption that users have some preference for locations close to the ones in which they take their photos. These virtual ratings help overcome the discrete nature of the geohashing. Second, we normalize the implicit frequency-based ratings to a 1-5 scale using a method that has been found to be useful in music recommendation algorithms. We demonstrate the advantages of our approach with new experiments that show large increases in hit rate and related metrics.