Practical recommender systems:
Saved in:
| Main Author: | |
|---|---|
| Format: | Book |
| Language: | English |
| Published: |
Shelter Island
Manning Publications
[2019]
|
| Subjects: | |
| Online Access: | Inhaltsverzeichnis Klappentext |
| Physical Description: | xxiv, 406 Seiten Illustrationen, Diagramme |
| ISBN: | 9781617292705 1617292702 |
Staff View
MARC
| LEADER | 00000nam a22000008c 4500 | ||
|---|---|---|---|
| 001 | BV044947740 | ||
| 003 | DE-604 | ||
| 005 | 20190919 | ||
| 007 | t | ||
| 008 | 180517s2019 a||| |||| 00||| eng d | ||
| 020 | |a 9781617292705 |9 978-1-61729-270-5 | ||
| 020 | |a 1617292702 |9 1-61729-270-2 | ||
| 035 | |a (OCoLC)1089721966 | ||
| 035 | |a (DE-599)BSZ486736504 | ||
| 040 | |a DE-604 |b ger |e rda | ||
| 041 | 0 | |a eng | |
| 049 | |a DE-355 | ||
| 084 | |a ST 515 |0 (DE-625)143677: |2 rvk | ||
| 100 | 1 | |a Falk, Kim |e Verfasser |0 (DE-588)1194693318 |4 aut | |
| 245 | 1 | 0 | |a Practical recommender systems |c Kim Falk |
| 264 | 1 | |a Shelter Island |b Manning Publications |c [2019] | |
| 300 | |a xxiv, 406 Seiten |b Illustrationen, Diagramme | ||
| 336 | |b txt |2 rdacontent | ||
| 337 | |b n |2 rdamedia | ||
| 338 | |b nc |2 rdacarrier | ||
| 650 | 0 | 7 | |a Suchmaschinenoptimierung |0 (DE-588)1026698669 |2 gnd |9 rswk-swf |
| 650 | 0 | 7 | |a Empfehlungssystem |0 (DE-588)7511891-9 |2 gnd |9 rswk-swf |
| 689 | 0 | 0 | |a Empfehlungssystem |0 (DE-588)7511891-9 |D s |
| 689 | 0 | 1 | |a Suchmaschinenoptimierung |0 (DE-588)1026698669 |D s |
| 689 | 0 | |5 DE-604 | |
| 856 | 4 | 2 | |m Digitalisierung UB Regensburg - ADAM Catalogue Enrichment |q application/pdf |u http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA |3 Inhaltsverzeichnis |
| 856 | 4 | 2 | |m Digitalisierung UB Regensburg - ADAM Catalogue Enrichment |q application/pdf |u http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000003&line_number=0002&func_code=DB_RECORDS&service_type=MEDIA |3 Klappentext |
| 999 | |a oai:aleph.bib-bvb.de:BVB01-030340532 | ||
Record in the Search Index
| _version_ | 1804178536720760832 |
|---|---|
| adam_text | contents preface xvii acknowUdgments xix about this book xx about the author xxiii about the cover illustration Part 1 Getting xxiv ready for recommender systems .... 1 Jį What is a recommender? 3 1.1 Real-life recommendations 3 Recommender systems are at home on the internet 5 ■ The long tail 5 ■ The Netflix recommender system 6 ■ Recommender system definition 12 1.2 Taxonomy of recommender systems 15 Domain 15 ■ Purpose 16 ■ Context 16 · Personalization level 17 · Whose opinions 18 ■ Privacy and trustworthiness 18 · Interface 19 ■ Algorithms 22 1.3 1.4 Machine learning and the Netflix Prize The MovieGEEKs website 24 Design and specification 1.5 26 · Architecture Building a recommender system IX 28 23 26
CONTENTS 2 User behavior and how to collect it 30 2.1 How (I think) Netflix gathers evidence while you browse 31 The evidence Netflix collects 2.2 33 Finding useful user behavior 35 Capturing visitor impressions 35 ■ What you can kam from a shop browser 36 · Act ofbuying 40 · Consuming products 41 Visitor ratings 42 · Getting to know your customers the (old) Netflix way 45 2.3 Identifying users 46 2.4 Getting visitor data from other sources 2.5 The collector 46 47 Building the project fiks 48 ■ The data model 48 The snitch: Client-side evidence colkctor 49 · Integrating the colkctor into MovieGEEKs 50 2.6 What users in the system are and how to model them 52 Jjl Monitoring the system 57 3.1 Why adding a dashboard is a good idea Answering “How are we doing?” 3.2 Doing the analytics 58 58 60 Web analytics 60 · The basic statistics 60 · Conversions Analyzing the path up to conversion 64 · Conversion path 3.3 Personas 68 3.4 MovieGEEKs dashboard 71 Auto-generating data to your log 71 ■ Specification and design of the analytics dashboard 72 ■ Analytics dashboard wireframe 72 ■ Architecture 73 ճէ. Ratings and how to calculate them 4.1 User-item preferences Definition of ratings 4.2 77 78 78 ■ User-item matrix Explicit or implicit ratings 81 How we use trusted sources for recommendations 4.3 Revisiting explicit ratings 83 4.4 What are implicit ratings? 83 Peopk suggestions ratings 85 79 82 85 ■ Considerations of calculating 61 66
CONTENTS 4.5 ХІ Calculating implicit ratings 88 Looking at the behavioral data 89· This could be considered a machine learning problem 93 4.6 How to implement implicit ratings Adding the time aspect 4.7 97 Less frequent items provide more value Non-personalized recommendations 5.1 102 103 103· What does a recommendation How to make recommendations when you have no data 105 Top 10: A chart of items 5.3 99 What’s a non-personalized recommendation? What’s a commerciali do? 105 5.2 93 107 Implementing the chart and the groundwork for the recommender system component 108 The recommender system component 108 · MovieGEEKs code from GitHub 110· A recommender system 110 · Adding a chart to MovieGEEKs 110· Making the content look more attractive 111 5.4 Seeded recommendations 113 Frequently bought items similar to the one you ’re viewing 114 Association rules 115· Implementing association rules 120 Saving the association rules in the database 123 · Running the association rules calculator 124· Using different events to create the association rules 126 The user (and content) who came in from the cold 6.1 What’s a cold start? 128 128 Cold products 130· A cold visitor 130· Gray sheep 132 Let’s look at real-life examples 132 · What can you do about cold starts? 133 6.2 Keeping track of visitors Persisting anonymous users 6.3 134 134 Addressing cold-start problems with algorithms 134 Using association rules to create recs for cold users 135 Using domain knowledge and business rules 136· Using segments 137· Using categories to get around the gray sheep problem and how to introduce cold
product 139
ХІІ CONTENTS 6.4 Those who doesn’t ask, won’t know When the visitor is no longer new 6.5 140 141 Using association rules to start recommending things fast 142 Find the coüected items 143 · Retrieve association rules and order them according to confidence 143 ■ Displaying the recs 144 ■ Implementation evaluation 147 Part 2 Recommender algorithms.............................149 y7 Finding similarities among users and among content 7.1 Why similarity? 152 What’s a similarity function ? 7.2 151 133 Essential similarity functions 153 Jaccard distance 153 · Measuring distance with Lp-norms 156 Cosine similarity 159■ Finding similarity with Pearson’s correlation coefficient 162 · Test running a Pearson similarity 163 * Pearson correlation is similar to cosine 165 7.3 к-means clustering 165 The h-means clustering algorithm clustering into itthon 168 7.4 Implementing similarities 166 · Translating k-means 172 Implementing the similarity in the MovieGEEKs site Implementing the clustering in the MovieGEEKs site Collaborative filtering in the neighborhood 8.1 Collaborative Filtering: A history lesson 174 177 181 183 When information became collaboratively filtered 183 Helping each other 183 ■ The rating matrix 185 ■ The collaborative filtering pipeline 186· Should you use user-user or item-item collaborativefiltering? 186 · Data requirements 187 8.2 8.3 8.4 8.5 8.6 8.7 8.8 Calculating recommendations 188 Calculating similarities 188 Amazon’s algorithm to precalculate item similarity Ways to select the neighborhood 194 Finding the right neighborhood 195 Ways to calculate predicted
ratings 196 Prediction with item-based filtering 197 Computing item predictions 198 189
XIU CONTENTS 8.9 8.10 8.11 Cold-start problems 199 A few words on machine learning terms 199 Collaborative filtering on the MovieGEEKs site Item-based, filtering 8.12 8.13 8.14 What’s the difference between association rule recs and collaborative recs? 207 Levers to fiddle with for collaborative filtering 207 Pros and cons of collaborative filtering 209 Evaluating and testing your recommender 9.1 9.2 9.3 9.4 200 202 211 Business wants lift, cross-sales, up-sales, and conversions 212 Why is it important to evaluate? 213 How to interpret user behavior 214 What to measure 214 Understanding my taste: Minimizing prediction error 216 Diversity 216 ■ Coverage 217 ■ Serendipity 219 9.5 Before implementing the recommender... Verify the algorithm 9.6 9.7 220 · Regression testing 219 221 Types of evaluation 222 Offline evaluation 222 What to do when the algorithm doesn’t produce any recommendations 223 9.8 Offline experiments 223 Preparing the data for the experiment 9.9 Implementing the experiment in MovieGEEKs 235 The to-do list 9.10 235 Evaluating the test set 239 Starting out with the baseline predictor parameters 242 9.11 Online evaluation 239· Finding the right 243 Controlkd experiments 9.12 229 243 ■ A/В testing 244 Continuous testing with exploit/explore Feedback loops 246 245
CONTENTS XIV Content-based filtering 248 10.1 10.2 Descriptive example 249 Content-based filtering 251 10.3 Content analyzer 253 Feature extraction for the item profik 253· Categorical data with small numbers 255 · Converting the year to a comparable feature 255 10.4 Extracting metadata from descriptions Preparing descriptions 10.5 10.6 256 256 Finding important words with TF-IDF Topic modeling using the LDA 261 260 What knobs can you turn to tweak the LDA ? 10.7 10.8 Finding similar content 271 Creating the user profile 272 Creating the user profile with LDA withTF-FDF 272 10.9 268 272 * Creating the user profile Content-based recommendations in MovieGEEKs 274 Loading data 274 · Training the model 277 · Creating item profiks 278 · Creating user profiks 278 · Showing recommendations 280 10.10 10.11 Evaluation of the content-based recommender Pros and cons of content-based filtering 282 Finding hidden genres with matrix factorization 11.1 11.2 11.3 284 Sometimes it’s good to reduce the amount of data Example of what you want to solve 287 A whiff of linear algebra 290 Matrix 11.4 281 290 · What’s factorization? 285 292 Constructing the factorization using SVD 293 Adding a new user by folding in 299 · How to do recommendations with SVD 301 · Baseline predictors Temporal dynamic 304 11.5 Constructing the factorization using Funk SVD 302 305 Root Mean Squared Error 305 · Gradient descent 306 Stochastic gradient descent 309 · And finally, to the factorization 309 · Adding biases 310 · How to start and when to stop 311
XV CONTENTS 11.6 11.7 Doing recommendations with Funk SVD 315 Funk SVD implementation in MovieGEEKs 318 What to do with outliers 322 · Keeping the model up to date 324 · Faster implementation 324 11.8 11.9 11.10 Explicit vs. implicit data 324 Evaluation 324 Levers to fiddle with for Funk SVD 326 Taking the best of all algorithms: Implementing hybrid recommenders 329 12.1 12.2 The confused world of hybrids The monolithic 331 330 Mixing content-based features with behavioral data to improve collaborative filtering recommenders 332 12.3 12.4 Mixed hybrid recommender 333 The ensemble 334 Switched ensemble recommender 335 · Weighted ensemble recommender 336 · Linear regression 337 12.5 Feature-weighted linear stacking (FWLS) Meta features: Weights as functions 12.6 Implementation 348 Ranking and learning to rank 13.1 13.2 13.3 357 Learning to rank an example at Foursquare 358 Re-ranking 362 What’s learning to rank again? 363 The three types of LTR algorithms 13.4 338 339 · The algorithm 363 Bayesian Personalized Ranking 365 Ranking with BPR 368 · Math magic (advanced wizardry) 369· The BPR algorithm 372 · BPR with matrix factorization 373 13.5 Implementation of BPR 373 Doing the recommendations 13.6 13.7 378 Evaluation 380 Levers to fiddle with for BPR 382 341
CONTENTS Future of recommender systems 14.1 14.2 This book in a few sentences Topics to study next 388 384 385 Further reading 388 ■ Algorithms 389 ■ Context 389 Human-computer interactions 390 ■ Choosing a good architecture 390 14.3 14.4 What’s the future of recommender systems? Final thoughts 395 index 397 391
SOFTWARE DEVELOPMENT/MACHINE LEARNING Practical Recommender Systems Kim Falk niine recommender systems help users find movies, jobs, restaurants—even romance! There’s an art in combin ing statistics, demographics, and query terms to achieve results that will delight them. Learn to build a recommender system the right way: it can make or break your application! í explains how recommender systems work and shows how to create and apply them for your site. After covering the basics, you’ll see how to collect user data and produce personalized recommendations. You’ll learn how to use the most popular recommendation algo rithms and see examples of them in action on sites like Ama zon and Netflix. Finally, the book covers scaling problems and other issues you’ll encounter as your site grows. f І՛ Լ .... ? Г„ • • • • ? . , Sn How to collect and understand üser behavior Collaborative and content-based filtering Machine learning algorithms Real-world examples in Python Readers need intermediate programming and database skills. is an experienced data scientist who works daily with machine learning and recommender systems. ՀՀ Covers the technical back ground and demonstrates implementations in clear and concise Python code. ^ —Andrew Collier, Exegetic ՀՀ Have you wondered how Amazon and Netflix learn your tastes in products and movies, and provide relevant recommendations? This book explains how it’s done!?? —Amit Lamba, Tech Overture ՀՀ Everything about recom mender systems, from entrylevel to advanced concepts.^ —Jaromir D.B. Němec, DBN ՀՀ A great and practical deep dive into
recommender systems! ^ —Peter Hampton Ulster University
|
| any_adam_object | 1 |
| author | Falk, Kim |
| author_GND | (DE-588)1194693318 |
| author_facet | Falk, Kim |
| author_role | aut |
| author_sort | Falk, Kim |
| author_variant | k f kf |
| building | Verbundindex |
| bvnumber | BV044947740 |
| classification_rvk | ST 515 |
| ctrlnum | (OCoLC)1089721966 (DE-599)BSZ486736504 |
| discipline | Informatik |
| format | Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01679nam a22003618c 4500</leader><controlfield tag="001">BV044947740</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">20190919 </controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">180517s2019 a||| |||| 00||| eng d</controlfield><datafield tag="020" ind1=" " ind2=" "><subfield code="a">9781617292705</subfield><subfield code="9">978-1-61729-270-5</subfield></datafield><datafield tag="020" ind1=" " ind2=" "><subfield code="a">1617292702</subfield><subfield code="9">1-61729-270-2</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)1089721966</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BSZ486736504</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-604</subfield><subfield code="b">ger</subfield><subfield code="e">rda</subfield></datafield><datafield tag="041" ind1="0" ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="049" ind1=" " ind2=" "><subfield code="a">DE-355</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">ST 515</subfield><subfield code="0">(DE-625)143677:</subfield><subfield code="2">rvk</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Falk, Kim</subfield><subfield code="e">Verfasser</subfield><subfield code="0">(DE-588)1194693318</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Practical recommender systems</subfield><subfield code="c">Kim Falk</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Shelter Island</subfield><subfield code="b">Manning Publications</subfield><subfield code="c">[2019]</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">xxiv, 406 Seiten</subfield><subfield code="b">Illustrationen, Diagramme</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Suchmaschinenoptimierung</subfield><subfield code="0">(DE-588)1026698669</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Empfehlungssystem</subfield><subfield code="0">(DE-588)7511891-9</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="689" ind1="0" ind2="0"><subfield code="a">Empfehlungssystem</subfield><subfield code="0">(DE-588)7511891-9</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2="1"><subfield code="a">Suchmaschinenoptimierung</subfield><subfield code="0">(DE-588)1026698669</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2=" "><subfield code="5">DE-604</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="m">Digitalisierung UB Regensburg - ADAM Catalogue Enrichment</subfield><subfield code="q">application/pdf</subfield><subfield code="u">http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA</subfield><subfield code="3">Inhaltsverzeichnis</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="m">Digitalisierung UB Regensburg - ADAM Catalogue Enrichment</subfield><subfield code="q">application/pdf</subfield><subfield code="u">http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000003&line_number=0002&func_code=DB_RECORDS&service_type=MEDIA</subfield><subfield code="3">Klappentext</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-030340532</subfield></datafield></record></collection> |
| id | DE-604.BV044947740 |
| illustrated | Illustrated |
| indexdate | 2024-07-10T08:05:32Z |
| institution | BVB |
| isbn | 9781617292705 1617292702 |
| language | English |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-030340532 |
| oclc_num | 1089721966 |
| open_access_boolean | |
| owner | DE-355 DE-BY-UBR |
| owner_facet | DE-355 DE-BY-UBR |
| physical | xxiv, 406 Seiten Illustrationen, Diagramme |
| publishDate | 2019 |
| publishDateSearch | 2019 |
| publishDateSort | 2019 |
| publisher | Manning Publications |
| record_format | marc |
| spelling | Falk, Kim Verfasser (DE-588)1194693318 aut Practical recommender systems Kim Falk Shelter Island Manning Publications [2019] xxiv, 406 Seiten Illustrationen, Diagramme txt rdacontent n rdamedia nc rdacarrier Suchmaschinenoptimierung (DE-588)1026698669 gnd rswk-swf Empfehlungssystem (DE-588)7511891-9 gnd rswk-swf Empfehlungssystem (DE-588)7511891-9 s Suchmaschinenoptimierung (DE-588)1026698669 s DE-604 Digitalisierung UB Regensburg - ADAM Catalogue Enrichment application/pdf http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA Inhaltsverzeichnis Digitalisierung UB Regensburg - ADAM Catalogue Enrichment application/pdf http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000003&line_number=0002&func_code=DB_RECORDS&service_type=MEDIA Klappentext |
| spellingShingle | Falk, Kim Practical recommender systems Suchmaschinenoptimierung (DE-588)1026698669 gnd Empfehlungssystem (DE-588)7511891-9 gnd |
| subject_GND | (DE-588)1026698669 (DE-588)7511891-9 |
| title | Practical recommender systems |
| title_auth | Practical recommender systems |
| title_exact_search | Practical recommender systems |
| title_full | Practical recommender systems Kim Falk |
| title_fullStr | Practical recommender systems Kim Falk |
| title_full_unstemmed | Practical recommender systems Kim Falk |
| title_short | Practical recommender systems |
| title_sort | practical recommender systems |
| topic | Suchmaschinenoptimierung (DE-588)1026698669 gnd Empfehlungssystem (DE-588)7511891-9 gnd |
| topic_facet | Suchmaschinenoptimierung Empfehlungssystem |
| url | http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000001&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=030340532&sequence=000003&line_number=0002&func_code=DB_RECORDS&service_type=MEDIA |
| work_keys_str_mv | AT falkkim practicalrecommendersystems |