Do you want to publish a course? Click here

Similarity calculation for recommender systems

حساب التشابه من أجل الأنظمة الناصحة

1216   1   80   0 ( 0 )
 Publication date 2014
and research's language is العربية
 Created by Shamra Editor




Ask ChatGPT about the research

Recommender systems represents a class of systems designed to help individuals deal with information overload or incomplete information. Such systems help individuals by providing recommendation through the use of various personalization techniques. Collaborative filtering is a widely used technique for rating prediction in recommender systems. This paper presents a method uses preference relations instead of absolute ratings for similarity calculation. The result indicates that the proposed method outperform the other methods such as the Somers Coefficient.


Artificial intelligence review:
Research summary
تتناول هذه الورقة البحثية موضوع الأنظمة الناصحة، وهي أنظمة مصممة لمساعدة الأفراد في التعامل مع المعلومات الزائدة أو الناقصة من خلال تقديم توصيات مخصصة. تركز الورقة على تقنية الفلترة التعاونية التي تُستخدم بشكل واسع في هذه الأنظمة لتحقيق التنبؤات. تقدم الباحثة طريقة جديدة لحساب التشابه بين المستخدمين باستخدام علاقات التفضيل بدلاً من التقييمات المطلقة، مما يساعد في تقديم توصيات أكثر دقة. تُظهر النتائج التجريبية أن الطريقة المقترحة تتفوق على طرق أخرى مثل معامل Somers. تم استخدام مجموعة بيانات من 443 مستخدم و840 سلعة لتحقيق الدراسة التجريبية، وتم مقارنة الطريقة المقترحة مع طريقة معامل Somers باستخدام مقياس متوسط الخطأ المطلق (MAE)، حيث أظهرت الطريقة الجديدة نتائج أفضل.
Critical review
دراسة نقدية: تقدم الورقة البحثية مساهمة قيمة في مجال الأنظمة الناصحة من خلال اقتراح طريقة جديدة لحساب التشابه بين المستخدمين. ومع ذلك، يمكن توسيع الدراسة لتشمل تقنيات أخرى مثل طريقة Pearson correlation وطريقة Random Walk Recommender لتحسين النتائج. كما يمكن دمج معطيات إضافية مثل البيانات الديموغرافية للمستخدمين لزيادة دقة التوصيات. بالإضافة إلى ذلك، قد يكون من المفيد اختبار الطريقة المقترحة على مجموعات بيانات أكبر وأكثر تنوعًا للتحقق من قابليتها للتعميم.
Questions related to the research
  1. ما هي التقنية الرئيسية المستخدمة في الأنظمة الناصحة لتحقيق التنبؤات؟

    التقنية الرئيسية المستخدمة هي الفلترة التعاونية.

  2. ما هي الطريقة الجديدة التي تقترحها الباحثة لحساب التشابه بين المستخدمين؟

    الطريقة الجديدة تعتمد على استخدام علاقات التفضيل بدلاً من التقييمات المطلقة.

  3. ما هو مقياس الأداء المستخدم لمقارنة الطريقة المقترحة مع الطرق الأخرى؟

    مقياس الأداء المستخدم هو متوسط الخطأ المطلق (MAE).

  4. ما هي الاقتراحات المستقبلية التي تقدمها الباحثة لتطوير الدراسة؟

    تقترح الباحثة توسيع الدراسة لتشمل تقنيات أخرى مثل طريقة Pearson correlation وطريقة Random Walk Recommender، وكذلك دمج معطيات ديموغرافية للمستخدمين.


References used
Desarkar M. et all., 2010 –Aggregating Preference Graphs for Collaborative Rating, RecSys, Proc. 4th ACM conference on Recommender systems.New York, USA
Chen Y.L. et all., 2008 – A novel collaborative filtering approach for recommending ranked items, Expert System with application
Jin R. et all., 2003 – Preference-based graphics models for collaborative filtering, UAI, page 329-336
Yildirim H., Kishnamoorthy M. S., 2008 – A random Walk method for alleviating the sparsity in collaborative filtering. RecSys, ACM, new York, USA, page 131 – 138
Moshfeghi D. et all., 2009 – Movie recommender: Sementically emrichedunified relevance model for rating predication in collaborative filtering, in ECIR page 54 – 65
rate research

Read More

Recommendation systems are the systems thathelp users to select suitable items from a large collection of items based on their tastes and interests. Such systems have become one of the most powerful tools in electronic commerce and social websites . Nonetheless , using these systems in e-commerce websites faces many drawbacks such as: cold start-up, scalability and sparsity. In this paper, we present a solution to cold-start-up problem, and compare between many association rule algorithms to select the most suitable one to solve the scalability and sparsity problems.
For many NLP applications of online reviews, comparison of two opinion-bearing sentences is key. We argue that, while general purpose text similarity metrics have been applied for this purpose, there has been limited exploration of their applicabilit y to opinion texts. We address this gap in the literature, studying: (1) how humans judge the similarity of pairs of opinion-bearing sentences; and, (2) the degree to which existing text similarity metrics, particularly embedding-based ones, correspond to human judgments. We crowdsourced annotations for opinion sentence pairs and our main findings are: (1) annotators tend to agree on whether or not opinion sentences are similar or different; and (2) embedding-based metrics capture human judgments of opinion similarity'' but not opinion difference''. Based on our analysis, we identify areas where the current metrics should be improved. We further propose to learn a similarity metric for opinion similarity via fine-tuning the Sentence-BERT sentence-embedding network based on review text and weak supervision by review ratings. Experiments show that our learned metric outperforms existing text similarity metrics and especially show significantly higher correlations with human annotations for differing opinions.
The thermal exchangers are widely used in the power generating stations and many refrigeration systems. Activity in the thermal cycle depends on its design and real calculations. The aim of our present work, is to find an ideal calculation method of the working exchangers parameters and evaluating the used technical engineering soluting. We discussed the ideal chosen problems and the affecting parameters, and used a calculating design for a thermal exchanger in the heating and feeding with hot water, considering these parameters. The proposed designing benefit, could be evaluated from the final indication that could be expressed by real palpable material. We found that the ideal parameters choosing could clearly descrese the preparation costs the heat exchanger blank – form by three times and give full clear proposed calculation method of all the parameters and its exchanging. This could save of the best exchanging possibility chosen from the different exchanger.
During the study of liquid filling systems’ performance using a piston driven by a pneumatic piston, all parameters having an influence should be considered. One of these parameters is the piston pressure and its influence on the velocity of the p iston. This influence will be explained in this paper.
Semantic textual similarity (STS) systems estimate the degree of the meaning similarity between two sentences. Cross-lingual STS systems estimate the degree of the meaning similarity between two sentences, each in a different language. State-of-the-a rt algorithms usually employ a strongly supervised, resource-rich approach difficult to use for poorly-resourced languages. However, any approach needs to have evaluation data to confirm the results. In order to simplify the evaluation process for poorly-resourced languages (in terms of STS evaluation datasets), we present new datasets for cross-lingual and monolingual STS for languages without this evaluation data. We also present the results of several state-of-the-art methods on these data which can be used as a baseline for further research. We believe that this article will not only extend the current STS research to other languages, but will also encourage competition on this new evaluation data.
comments
Fetching comments Fetching comments
Sign in to be able to follow your search criteria
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا