Integration of Linked Open Data in Collaborative Group Recommender Systems

A Recommender System (RS) is an intelligent system that assists users in finding the items of their interest (e.g. books, movies, music) by preventing them to go through huge piles of data available online. In an effort to overcome the data sparsity issue in recommender systems, this research incorporates a content based filtering technique with fuzzy inference system and a conformal prediction approach introducing a new framework called Hybrid Content based Fuzzy Conformal Recommender System (HCF-CRS). The proposed framework is implemented to be used in the domain of movies and it provides quality recommendations to users with a confidence level and an improved accuracy. In our proposed framework, first, a Content Based Filtering (CBF) technique is applied to create a user profile by considering the history of each user. CBF is useful in the situations like: lack of demographic information and the data sparsity problems. Second, a Fuzzy based technique is incorporated to find the similarities and differences between the user profile and the movies in the dataset using a set of fuzzy rules to get a predicted rating for each movie. Third, a Conformal prediction algorithm is implemented to calculate the non-conformity measure between the predicted ratings produced by fuzzy system and the actual ratings from the dataset. A p-value (confidence measure) is computed to give a level of confidence to each recommended item and a bound is set on the confidence level called a significance level ε, according to which the movies only above the specified significance level are recommended to user. By building a confidence centric hybrid conformal recommender system using the content based filtering approach with fuzzy logic and conformal prediction algorithm, the reliability and the accuracy of the system is considerably enhanced. The experiments are evaluated on MovieLens and Movie Tweetings datasets for recommending movies to the users and they are compared with other state-of-the-art recommender systems. Finally, the results confirm that the proposed algorithms perform better than the traditional ones.

The conventional seismic inversion approach is practical for operational work, as it only uses simple linearized algorithms and assumptions, but may be less applicable when dealing with a complex geological setting, especially in the Malay basin fields, as it may introduce non-linear noises and non-unique solutions. In the Malay basin, we also frequently struggle with a scarcity of reliable well data when performing seismic inversions. This makes finding an accurate prior model for inversion challenging and contributes to high uncertainty in properties’ estimation. Implementation of deep learning for seismic inversion has become routine and has shown increasing capability in addressing nonlinearity in inverse problems. In this work, we develop a robust approach to deep learning-based seismic inversion to predict elastic properties from seismic data. The approach incorporates synthetic well and seismic data generation from a set of rock physics knowledge called the rock physics library, which plays a significant role in dataset input for network training, validation, and testing to improve elastic properties in this field. The deep learning network architecture comprising UNET and RESNET-18 with weak supervision networks has proven to be useful to enhance computational work efficiency and prediction accuracy while handling the non-linearity of the data and the non-uniqueness of the solutions. We successfully validated the proposed method on actual field data from a clastic fluvial-dominated field in the Malay basin. Upon comparative analysis with the conventional method, both inversion results are comparable and capable of identifying the reservoir occurrence and distribution. The conventional method exposed the presence of scattered amplitude noises and prominent seismic imprints masking the reservoir. Meanwhile, the proposed method showed more stable, clearer definition and fewer noise inversion results but with a faster turn-around time and a more efficient workflow. There are substantial improvements of up to 31% in correlation accuracy achieved upon implementing the proposed method for elastic properties prediction compared to the conventional. The result implies that the proposed method can provide a good elastic properties prediction framework while addressing data limitations and sparsity issues in typical deep learning-based inversions.

Reproducibility is a challenging aspect that considerably affects the quality of most scientific papers. To deal with this, many open frameworks allow to build, test, and benchmark recommender systems for single users. Group recommender systems involve additional tasks w.r.t. those for single users, such as the identification of the groups, or their modeling. While this clearly amplifies the possible reproducibility issues, to date, no framework to benchmark group recommender systems exists. In this work, we enable reproducibility in group recommender systems by extending the LibRec library, which stands out as one of the richest, with more than 70 different recommender algorithms, good performance and several evaluation metrics. Specifically, we include several approaches for all the stages of group recommender systems: group formation, group modeling strategies, and evaluation. To validate our framework, we consider a use-case that compares several group building, recommendation, and group modeling approaches.

Group size influences many aspects of mammalian social life, including stress levels, disease transmission, reproductive rates, and behavior. However, much of what is known about the effects of group size on behavioral ecology has come from comparisons across multiple groups of different sizes. These findings may be biased because behavioral differences across groups may be more indicative of how environmental variation influences animal behavior, rather than group size itself. To partially circumvent this limitation, we used longitudinal data to examine how changes in group size across time affect the behavior of folivorous red colobus monkeys (Procolobus rufomitratus) of Kibale National Park, Uganda. Controlling for food availability, we demonstrated that increasing group size resulted in altered activity budgets, based on 6 yr of data on a group that increased from 57 to 98 members. Specifically, as group size increased, individuals spent less time feeding and socializing, more time traveling, and increased the diversity of their diet. These changes appear to allow the monkeys to compensate for greater scramble competition apparent at larger group sizes, as increasing group size did not show the predicted relationship with lower female fecundity. Our results support recent findings documenting feeding competition in folivorous primates. Our results also document behavioral flexibility, an important trait that allows many social mammals to maximize the benefits of sociality (e.g., increased vigilance), while minimizing the costs (e.g., increased feeding competition).

For decades, scholars have been trying to determine whether small or large groups are more likely to cooperate for collective action and successfully manage common-pool resources. Using data gathered from the Wolong Nature Reserve since 1995, we examined the effects of group size (i.e., number of households monitoring a single forest parcel) on both collective action (forest monitoring) and resource outcomes (changes in forest cover) while controlling for potential confounding factors. Our results demonstrate that group size has nonlinear effects on both collective action and resource outcomes, with intermediate group size contributing the most monitoring effort and leading to the biggest forest cover gain. We also show how opposing effects of group size directly and indirectly affect collective action and resource outcomes, leading to the overall nonlinear relationship. Our findings suggest why previous studies have observed differing and even contradictory group-size effects, and thus help guide further research and governance of the commons. The findings also suggest that it should be possible to improve collective action and resource outcomes by altering factors that lead to the nonlinear group-size effect, including punishing free riding, enhancing overall and within-group enforcement, improving social capital across groups and among group members, and allowing self-selection during the group formation process so members with good social relationships can form groups autonomously.

The fact that most female primates (and many other mammals) live in groups is paradoxical, given that the presence of others presumably increases competition for foods and may, for some, reduce reproductive success. Competition for food resources is generally inferred from any of the following observations: (1) female dominance hierarchies within groups; (2) female aggression between groups; (3) increasing home-range size with increasing group size; (4) longer day-range length with increasing group size; and (5) lower reproductive rates in larger groups. Both female aggression (interference competition) and adjustments of ranging behavior to group size (exploitative competition) have been linked in the past to patterns of food distribution and abundance. Using data largely from the literature, this paper examines the covariance of female aggression and ranging behavior among 20 species of primates in an attempt to better explain the variation in female relationships within and between groups of primates. Results show that groups of females are aggressive toward other groups and that home-range size increases with increasing group size in most species. In addition, in those species with strong dominance hierarchies within groups, day-range length increases as a function of group size. However, in those species that do not have strong dominance hierarchies within groups, dayrange length does not increase as a function of group size. The implications of these results are presented in a model that suggests that intergroup competition is determined by food abundance, whereas intragroup competition is determined by food distribution. This model differs from earlier models in its explanation of the ecological conditions that influence female relationships within and between groups of primates. [Behav Ecol 1991;2:143–155]

Six group sizes and three levels of floor area allowance were studied in a 6 x 3 factorial arrangement. Group sizes were 3, 5, 6, 7, 10, and 15 pigs per pen. Floor area allowances were .030, .039, and .048 m2 x BW(.667). All pens were square and equipped with a single space feeder and nipple drinker near one corner. Pigs were fed a pelleted diet. Initial weights averaged 25.0 kg, and pigs remained on test for 12 wk. Pigs were weighed, feed intake was determined, and size of pens was increased at 2-wk intervals. Pen size was adjusted to provide the space required for the midpoint of each weigh period. Two replicates of the study were conducted. The ADG was reduced with increasing group size (899, 851, 868, 872, 857, and 821 g, SEM = 16.4, for 3, 5, 6, 7, 10, and 15 pigs, respectively; P < .05), but the CV for gain (mean = .185) did not differ among group sizes. The ADFI also decreased with increasing group size (2.49, 2.34, 2.32, 2.28, 2.28, and 2.21 kg, SEM = .036, for 3, 5, 6, 7, 10, and 15 pigs, respectively; P < .05). Feed efficiency (gain/intake) was highest for group sizes of 7 and 10 pigs (.381) and lowest for pens of three and five pigs (.363; P < .05). The ADG and ADFI (832 g and 2.25 kg, respectively) for the most crowded space allowance were reduced compared to more spacious allowances (ADG and ADFI of 875 and 877 g, and 2.35 and 2.36 kg, for .039 and .048 m2 x BW(.667), respectively; P < .05). Efficiency did not differ among space allowances. These results confirm previous studies reporting a negative effect of increasing group size on productivity, but our study suggests that gain and intake reach a plateau at less space allowance than previously reported.

In socially feeding birds and mammals, as group size increases, individuals devote less time to scanning their environment and more time to feeding. This vigilance “group size effect” has long been attributed to the anti-predatory benefits of group living, but many investigators have suggested that this effect may be driven by scramble competition for limited food. We addressed this issue of causation by focusing on the way in which the scan durations of free-living dark-eyed juncos (Junco hyemalis) decrease with group size. We were particularly interested in vigilance scanning concomitant with the handling of food items, since a decrease in food handling times (i.e. scan durations) with increasing group size could theoretically be driven by scramble competition for limited food resources. However, we showed that food-handling scan durations decrease with group size in an environment with an effectively unlimited food supply. Furthermore, this food-handling effect was qualitatively similar to that observed in the duration of standard vigilance scans (scanning exclusive of food ingestion), and both responded to changes in the risk of predation (proximity of a refuge) as one might expect based upon anti-predator considerations. The group size effects in both food-handling and standard scan durations may reflect a lesser need for personal information about risk as group size increases. Scramble competition may influence vigilance in some circumstances, but demonstrating an effect of competition beyond that of predation may prove challenging.

Trust prediction in online social networks is crucial for information dissemination, product promotion, and decision making. Existing work on trust prediction mainly utilizes the network structure or the low-rank approximation of a trust network. These approaches can suffer from the problem of data sparsity and prediction accuracy. Inspired by the homophily theory, which shows a pervasive feature of social and economic networks that trust relations tend to be developed among similar people, we propose a novel deep user model for trust prediction based on user similarity measurement. It is a comprehensive data sparsity insensitive model that combines a user review behavior and the item characteristics that this user is interested in. With this user model, we firstly generate a user's latent features mined from user review behavior and the item properties that the user cares. Then we develop a pair-wise deep neural network to further learn and represent these user features. Finally, we measure the trust relations between a pair of people by calculating the user feature vector cosine similarity. Extensive experiments are conducted on two real-world datasets, which demonstrate the superior performance of the proposed approach over the representative baseline works.

Group Recommendation (GR) is the task of suggesting relevant items/events for a group of users in online systems, whose major challenge is to aggregate the preferences of group members to infer the decision of a group. Prior group recommendation methods applied predefined static strategies for preference aggregation. However, these static strategies are insufficient to model the complicated decision making process of a group, especially for occasional groups which are formed adhoc. Compared to conventional individual recommendation task, GR is rather dynamic and each group member may contribute differently to the final group decision. Recent works argue that group members should have non-uniform weights in forming the decision of a group, and try to utilize a standard attention mechanism to aggregate the preferences of group members, but they do not model the interaction behavior among group members, and the decision making process is largely unexplored.In this work, we study GR in a more general scenario, that is Occasional Group Recommendation (OGR), and focus on solving the preference aggregation problem and the data sparsity issue of group-item interactions. Instead of exploring new heuristic or vanilla attention-based mechanism, we propose a new social self-attention based aggregation strategy by directly modeling the interactions among group members, namely Group Self-Attention (GroupSA). In GroupSA, we treat the group decision making process as multiple voting processes, and develop a stacked social self-attention network to simulate how a group consensus is reached. To overcome the data sparsity issue, we resort to the relatively abundant user-item and user-user interaction data, and enhance the representation of users by two types of aggregation methods. In the training process, we further propose a joint training method to learn the user/item embeddings in the group-item recommendation task and the user-item recommendation task simultaneously. Finally, we conduct extensive experiments on two real-world datasets. The experimental results demonstrate the superiority of our proposed GroupSA method compared to several state-of-the-art methods in terms of HR and NDCG.

The data sparsity problem is addressed by using the decision tree state clusters as the training targets for the state-of-the-art context-dependent (CD) deep neural network (DNN) systems. The CD states within a cluster cannot be distinguished at the frame level. We surmise that the state clustering may cause an issue for the standard CD-DNNs, which has so far not been addressed in the literature. In this paper, a logistic regression framework is proposed for the CD-DNNs based on a set of broad phone classes to address both the data sparsity and the clustering problems. To address the data sparsity issue, the triphones are clustered into shorter biphones with broad phone contexts under multiple articulatory categories. A DNN is trained to discriminate the disjoint biphone clusters within each articulatory category. The regression bases are formed by the concatenated log posterior probabilities of all the broad phone DNNs. Logistic regression is used to transform the regression bases into the triphone state posteriors. Clustering of the regression parameters is used to reduce the regression model complexity while still achieving unique acoustic scores for all possible triphones. Based on some approximations, the regression model can be trained as a sparse softmax layer and its parameters can be learned by optimizing the cross-entropy criterion. The experimental results on a broadcast news transcription task reveal that the proposed regression-based CD-DNN significantly outperforms the standard CD-DNN. The best system provides a 1.3% absolute word error rate reduction compared to the best standard CD-DNN system.

1. Supplied with ad libitum prey, the effects of different group sizes (one, five, ten, twenty‐five, fifty or 100) on the acquisition and allocation of energy were examined over the entire life cycle of the freshwater predatory leech Nephelopsis obscura, to evaluate the bioenergetic costs and/or benefits of aggregation.2. In terms of growth, asymptotic biomass and ingestion, group sizes larger than one and less than ten are optimal. As group size increases from ten to 100, N. obscura has slower growth and lower weekly food ingestion, suggesting interference competition. Single leeches had lower food consumption than group sizes of less than 100 animals. With a lower asymptotic biomass, single leeches have lower fecundity and presumed lower fitness. Lipid deposition was unaffected by group size.3. Size‐ and time‐specific respiration rates declined with increasing group size. For group sizes between five and fifty, the same relative proportion of ingested energy was allocated to respiration. Above this threshold group size, leeches allocated a higher proportion of ingested energy to growth, i.e. had higher growth efficiency, as did leeches cultured individually.4. In the largest group size, total time spent in activity was highest (reflecting the high degree of interference between conspecifics), but as a result of the lower hourly costs of respiration they allocated a lower proportion of ingested energy to total respiration.5. Optimum group size of N. obscura in the laboratory is within the typical range of group sizes found on stony shores in the field.

Novel automatic group identification approaches for group recommendation

Decreasing aggression with increasing group size in young domestic fowl

In The Logic of Collective Action, Mancur Olson shows how the activities of various political organizations can be fruitfully analyzed using the theory of collective goods. Several of Olson's major conclusions concern the relationship between the size of a group and its ability to provide its members with collective benefits. He concludes that as group size increases, the amount of collective benefits provided will become increasingly suboptimal, and that the absolute amount of collective benefits provided will decrease. This paper shows that Olson's conclusion concerning the relationship between group size and the absolute amount of collective benefits provided is not generally true. Within the framework of analysis used by Olson, it is demonstrated that the relationship is determined by the interaction between two effects, an “income” effect which may cause the level of benefits provided to increase as group size increases, and a “congestion” effect which may cause the level of benefits to decrease as group size increases. The final result is that for “inclusive” collective goods the relationship is an increasing one, while for “exclusive” goods it is a decreasing one. Some implications of this result for the use of the theory of collective goods in studying political processes are discussed.