Cascade Hybrid Recommendation as a Combination of One-Class Classification and Collaborative Filtering

2014 ◽  
Vol 23 (04) ◽  
pp. 1460009 ◽  
Author(s):  
Aristomenis S. Lampropoulos ◽  
Dionisios N. Sotiropoulos ◽  
George A. Tsihrintzis
Keyword(s):  
Collaborative Filtering ◽  
Classification Scheme ◽  
Mean Absolute Error ◽  
Absolute Error ◽  
Case Scenario ◽  
Worst Case ◽  
One Class Classifier ◽  
The One ◽  
One Class Classification ◽  
Hybrid Recommendation

In this paper, we formulate the recommendation problem as a hybrid combination of one-class classification with collaborative filtering. Specifically, we decompose the recommendation problem into a two-level cascade scheme. In the first level, only desirable items are selected for each user from the large amount of all possible items, taking into account only a small portion of his/her available preferences. This is achieved via a one-class classification scheme trained only with positives examples, i.e. only with desirable items for which users have provided a rating value. In the second level, a collaborative filtering approach is applied to assign a rating degree to the items identified at the first level. The efficiency of our approach is analyzed theoretically in terms of best/worst case scenarios and respective lower/upper mean absolute error (MAE) bounds are computed. Moreover, our approach is experimentally tested against pure collaborative and cascade content-based approaches. The results show that our approach outperforms them in terms of MAE and, moreover, the experimental MAE is close to the theoretical lower bound corresponding to the best case scenario. The superiority of our approach is due to the existence of the one class classifier in the first level of the cascade.

Optimized Collaborative Filtering Recommendation Based on User’s Social Relationships

2014 ◽  
Vol 1044-1045 ◽  
pp. 1484-1488
Author(s):  
Yue Kun Fan ◽  
Xin Ye Li ◽  
Meng Meng Cao
Keyword(s):  
Collaborative Filtering ◽  
Social Relationships ◽  
Digital Libraries ◽  
Social Relationship ◽  
Nearest Neighbor ◽  
Mean Absolute Error ◽  
Absolute Error ◽  
Personalized Recommendation ◽  
Recommendation Algorithm

Currently collaborative filtering is widely used in e-commerce, digital libraries and other areas of personalized recommendation service system. Nearest-neighbor algorithm is the earliest proposed and the main collaborative filtering recommendation algorithm, but the data sparsity and cold-start problems seriously affect the recommendation quality. To solve these problems, A collaborative filtering recommendation algorithm based on users' social relationships is proposed. 0n the basis of traditional filtering recommendation technology, it combines with the interested objects of user's social relationship and takes the advantage of the tags to projects marked by users and their interested objects to improve the methods of recommendation. The experimental results of MAE ((Mean Absolute Error)) verify that this method can get better quality of recommendation.

Topology Optimization of Dynamic Systems Under Uncertain Loads: An H∞-Norm-Based Approach

2019 ◽  
Vol 14 (2) ◽  
Author(s):  
Paolo Venini
Keyword(s):  
Topology Optimization ◽  
Structural Parameters ◽  
Goal Function ◽  
Dynamic Compliance ◽  
Case Scenario ◽  
Worst Case ◽  
Original Definition ◽  
Optimal Material ◽  
The One ◽  
Definition Of

An innovative approach to topology optimization of dynamic system is introduced that is based on the system transfer-function H∞-norm. As for the structure, the proposed strategy allows to determine the optimal material distribution that ensures the minimization of a suitable goal function, such as (an original definition of) the dynamic compliance. Load uncertainty is accounted for by means of a nonprobabilistic convex-set approach (Ben-Haim and Elishakoff, 1990, Convex Models of Uncertainty in Applied Mechanics, Elsevier Science, Amsterdam). At each iteration, the worst load is determined as the one that maximizes the current dynamic compliance so that the proposed strategy fits the so-called worst case scenario (WCS) approach. The overall approach consists of the repeated solution of the two steps (minimization of the dynamic compliance with respect to structural parameters and maximization of the dynamic compliance with respect to the acting load) until convergence is achieved. Results from representative numerical studies are eventually presented along with extensions to the proposed approach that are currently under development.

scholarly journals Intelligent design af fokusgrupper - om metodisk design af fokusgrupper og menneskets forskellige intelligenser

2008 ◽  
Vol 24 (45) ◽  
pp. 13
Author(s):  
Lene Heiselberg
Keyword(s):  
Focus Groups ◽  
Intelligent Design ◽  
Verbal Intelligence ◽  
Case Scenario ◽  
Worst Case ◽  
Apparent Lack ◽  
Self Confidence ◽  
Media Research ◽  
The One ◽  
Interviewing Techniques

Når man arbejder professionelt med at gennemføre kvalitative mini- og fokusgruppeanalyser, kan det ikke undgås, at man som moderator indimellem tænker: Hvorfor deltager hun ikke? Hvad kan jeg gøre for at inkludere hende i diskussionen? Ofte skyldes nogle deltageres manglende engagement, at mini- eller fokusgruppens metodiske design favoriserer de deltagere, som har en fremtrædende verbalsproglig intelligens, og samtidig ekskluderes de, der har andre fremtrædende intelligenser, fra at yde det maksimale. En sådan situation er meget uheldig og kan i værste fald give en undersøgelse bias. Derfor har vi i DR Medieforskning arbejdet med en pragmatisk tilgang til problemet, hvor vi har afprøvet et metodisk design, som inkluderer kvalitative interviewteknikker og procesværktøjer, som appellerer til samtlige intelligenser. Som et resultat af en målrettet indsats for at inkludere flere intelligenser i det metodiske design, oplever vi, at deltagerne har mere lyst til at engagere sig og gør det med større selvsikkerhed. Desuden oplever vi i mindre grad fænomenet “cognitive tuning” , og derfor kan vi arbejde med flere og bedre data i analyse- og fortolkningsfasen. Intelligent design of focus groups - article about methodological design of focus groups and the different intelligences When you work professionally with the conducting and moderating of qualitative mini- and focus groups, you can't avoid sometimes thinking: Why isn’t she participating? What can I do to include her in the discussion? A participant's apparent lack of enthusiasm is often caused by the methodological design of the focus group giving preference to participants who have an explicit verbal intelligence, and as a consequence excludes participants with other explicit intelligences from contributing. A situation like the one described above is very undesirable and in a worst-case scenario it can cause a study to be biased. In order to try to solve this problem DR Media Research applied a methodological design which includes qualitative interviewing techniques and processing tools, which appeal to all of the intelligences instead of just one. As a result of this work, we find that the participants are more eager to participate and that they do it with greater self-confidence. In addition we encounter less cognitive tuning, and are therefore able to work with richer data in the phases of analysis and interpretation.

scholarly journals Pembuatan Modul Rekomendasi pada OpenCart Menggunakan Metode Item-Based Collaborative Filtering

2013 ◽  
Vol 2 (1) ◽  
pp. 9
Author(s):  
Kirana Nuryunita ◽  
Yani Nurhadryani
Keyword(s):  
Collaborative Filtering ◽  
Recommender System ◽  
Management System ◽  
Mean Absolute Error ◽  
Absolute Error ◽  
Content Management ◽  
Cosine Similarity ◽  
Weighted Sum ◽  
Content Management System

<p>Penelitian ini bertujuan menambahkan modul rekomendasi pada content management system Opencart. Salah satu pendekatan dalam rekomendasi adalah item-based collaborative filtering. Metode item-based collaborative filtering dapat mengurangi waktu eksekusi perhitungan. Metode item-based collaborative filtering pada penelitian ini menggunakan perhitungan adjusted cosine similarity untuk menghitung nilai kemiripan antarbuku dan weighted sum untuk menghitung nilai prediksi rate buku. Untuk mendapatkan rekomendasi, pengguna harus melakukan login dan memberikan rate pada buku. Berdasarkan rate pengguna, nilai kemiripan dihitung menggunakan adjusted cosine similarity. Berdasarkan kemiripan antarbuku, nilai prediksi rate buku dicari menggunakan weighted sum. Sebelum buku direkomendasikan kepada pengguna, kategori prediksi buku dicocokkan dengan kategori buku yang telah diberi rate oleh pengguna. Penelitian ini menggunakan 300 buku dan 30 pengguna sebagai data. Dari hasil penelitian, hanya 17 pengguna yang mendapatkan rekomendasi. Pengujian dilakukan dengan menganalisis waktu eksekusi dan keakuratan rekomendasi. Waktu eksekusi dalam pengujian ini adalah 1.60 detik. Untuk menghitung keakuratan rekomendasi, penelitian ini menggunakan mean absolute error dengan hasil perhitungan 0.15.</p><p>Kata kunci: e-commerce, item-based collaborative filtering, recommender system.</p>

China positions itself to exploit 'America First'

2017 ◽  
Keyword(s):  
South China Sea ◽  
South China ◽  
North Korea ◽  
The South China Sea ◽  
Strategic Partnership ◽  
Case Scenario ◽  
China Policy ◽  
Worst Case ◽  
Content Type ◽  
The One

Significance Trump affirmed US commitment to the 'one-China policy' vis-à-vis Taiwan, which he had questioned following his election. Nevertheless, Trump's willingness to raise the issue in the first place -- and his other post-election comments on North Korea and the South China Sea -- lead Beijing to expect an unprecedentedly rocky relationship with Washington during his term. Impacts Trump seems now to accept that questioning the one-China policy is taboo, but he could still provoke Beijing regarding Tibet. The combination of uncertain US policy and a China-sceptic government in Taipei will prompt Chinese preparations for a worst-case scenario. US-Russia rapprochement could complicate Beijing's strategic partnership with Moscow. Other governments stand to benefit from a Chinese 'charm offensive', as Beijing attempts to win friends rather than confront Washington.

scholarly journals Digitization: Does It Always Improve Access to Rare Books and Special Collections?

2017 ◽  
Vol 45 (4) ◽  
pp. 177-179 ◽  
Author(s):  
Dale J. Correa
Keyword(s):  
Middle East ◽  
Case Scenario ◽  
Rare Book ◽  
Worst Case ◽  
Special Collections ◽  
State Archives ◽  
The One ◽  
Physical Access ◽  
Conflict Research ◽  
Middle East Studies

Abstract:In April 2016, the Israel State Archives announced the most recent stage of an ambitious project to digitize all of their holdings (potentially 400 million pages of material): the new archival website was ready online (Aderet). With the new website came the ability to request digital copies of documents, which would be available on the website within two weeks of the request (Aderet). However, researchers would now at the very least be discouraged from requesting access to the paper documents (Lozowick), or, in the worst-case scenario, be refused access to anything except the website (Baron and Newhall). Local scholars (including a prominent professor of history at Tel Aviv University), the Akevot Institute for Israeli-Palestinian Conflict Research, and the Middle East Studies Association of North America (which publishes the prestigious International Journal of Middle East Studies) registered concern with the restriction of physical access to the archive and issued public calls for a reversal of the decision (Akevot Institute). The conflict was between perceived best practices of digitization and of archival stewardship (represented by the State Archivist Dr. Yaacov Lozowick) on the one hand, and standards and expectations for scholarly research on the Middle East, which largely depends on archival and rare book collections, on the other.

scholarly journals Modelling future trajectories of obesity and body mass index in England

PLoS ONE ◽  
2021 ◽  
Vol 16 (6) ◽  
pp. e0252072
Author(s):  
Linda J. Cobiac ◽  
Peter Scarborough
Keyword(s):  
Body Mass Index ◽  
Body Mass ◽  
Linear Models ◽  
Absolute Error ◽  
Outcome Variable ◽  
Case Scenario ◽  
Worst Case ◽  
Worst Case Scenario ◽  
Increased Risk ◽  
Non Linear

Background Obesity is a leading risk for poor health outcomes in England. We examined best- and worst-case scenarios for the future trajectory of the obesity epidemic. Methods Taking the last 27 years of Health Survey for England data, we determined both position and shape of the adult body mass index (BMI) distribution and projected these parameters 20 years forward in time. For the best-case scenario, we fitted linear models, allowing for a quadratic relationship between the outcome variable and time, to reflect a potential reversal in upwards trends. For the worst-case scenario, we fitted non-linear models that applied an exponential function to reflect a potential flattening of trends over time. Best-fitting models were identified using Monte Carlo cross-validation on 1991–2014 data, and predictions of population prevalence across five BMI categories were then validated using 2015–17 data. Results Both linear and non-linear models showed a close fit to observed data (mean absolute error <2%). In the best-case scenario, the proportion of the population at increased risk (BMI≥25kg/m2) is predicted to fall from 66% in 2017 to 53% (95% confidence interval: 41% to 64%) in 2035. In the worst-case scenario, this proportion is likely to remain relatively stable overall– 64% (37% to 90%) in 2035 –but with an increasing proportion of the population at highest risk (BMI≥35kg/m2). Conclusions While obesity prediction depends on chosen modelling methods, even under optimistic assumptions it is likely that the majority of the English population will still be at increased risk of disease due to their weight until at least 2035, without greater allocation of resources to effective interventions.

scholarly journals Being an Italian medical student during the Covid-19 outbreak

2020 ◽  
Author(s):  
Nicolo Giuseppe Biavardi
Keyword(s):  
Medical Student ◽  
Case Scenario ◽  
Academic World ◽  
Worst Case ◽  
Worst Case Scenario ◽  
The World ◽  
E Learning ◽  
The One

Many students around the world have been wondering how their life will change since the very first outbreak of Covid-19. In my experience article I have tried to give a flavor of how has the academic world changed in quarantine. Difficulties and opportunities have been analyzed. Questions regarding the validity of e-learning have been posed. In an arduous period as the one we are experiencing, having an idea of what life could be in worst case scenario could be helpful.

Sign in / Sign up

Export Citation Format

Share Document