Re: implementation of context-aware recommender in Mahout
The new Spark based recommender can easily handle context in many forms. See the top references section here http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html It does not use the IDRescorer approach at all so perhaps you should describe what you want to use as context. In the demo site for the new stuff (a guide to online video) https://guide.finderbots.com you’ll see a couple examples of “context”. For instance you are viewing a video that has several genre tags. You’ll see at least 3 lists of recommendations: 1) people who like the video you are looking at also like these other viedeos—non-personalized recs 2) people who like this video liked these, from similar genres 3) personalized recs from all genres based on your “liking” history Many other things can be used as context like time of day, location, mobile or desktop, user profile attributes, etc. The way it does this is through the search engine, which can take filters and boost certain item attributes. So I could show only recommendations made in the same year as the viewed movie or use the year to bias recommendations by boosting the “release-date” field in the recommender query. The recommender is also multimodal and so can use many user actions to better the quality of recs. Removing some of your data, in what you call pre-filtering may not get you what you want. Removing data that is actual user behavior can reduce the quality of recommendations so please give an example. On Mar 6, 2015, at 4:45 AM, Efi Koulouri wrote: Hi all, I am trying to implement an context-aware recommender in Mahout. As I haven't use the library before I haven't a lot experience. So, I would really appreciate your response! What I want to do is to implement the two context- aware approaches that have been proposed, pre-filtering and post-filtering. The former filters out the dataset based on the value of contextual factor before the collaborative filtering while the latter rescores the recommendations after the collaborative filtering. I have already read older similar questions regarding the context-aware recommender implementation in mahout and I know that the post-filtering method can be implemented using the IDRescorer. For the pre-filtering approach there is the option to use the CandidateItemsStategy in case of the item-based recommender. On the other hand if we want to implement this approach using the user-bsed recommender no such option is available. In order to implement the pre-filtering using the user-based recommender, I was thinking to filter out the unrelated user,items pairs from the dataset before the creation of the data model. This means that the data model will take as input a subset of the initial dataset. Does this approach sound correct? There are some concerns regarding the evaluation of the recommender. Does it have any impact on this? Thank you in advance! Regards, Efi
Re: implementation of context-aware recommender in Mahout
Thanks for your reply! Actually, the context in pre-filtering serves as a query for selecting relevant data. An example of a contextual data filter for a movie recommender system would be: if a person wants to see a movie on Saturday, only the Saturday rating data is used to recommend movies. So,what I need for the rating prediction is the data relevent to the specific context. Regards, Efi On 7 March 2015 at 01:32, Pat Ferrel wrote: > The new Spark based recommender can easily handle context in many forms. > See the top references section here > http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html > > It does not use the IDRescorer approach at all so perhaps you should > describe what you want to use as context. > > In the demo site for the new stuff (a guide to online video) > https://guide.finderbots.com you’ll see a couple examples of “context”. > For instance you are viewing a video that has several genre tags. You’ll > see at least 3 lists of recommendations: > 1) people who like the video you are looking at also like these other > viedeos—non-personalized recs > 2) people who like this video liked these, from similar genres > 3) personalized recs from all genres based on your “liking” history > > Many other things can be used as context like time of day, location, > mobile or desktop, user profile attributes, etc. The way it does this is > through the search engine, which can take filters and boost certain item > attributes. So I could show only recommendations made in the same year as > the viewed movie or use the year to bias recommendations by boosting the > “release-date” field in the recommender query. The recommender is also > multimodal and so can use many user actions to better the quality of recs. > > Removing some of your data, in what you call pre-filtering may not get you > what you want. Removing data that is actual user behavior can reduce the > quality of recommendations so please give an example. > > On Mar 6, 2015, at 4:45 AM, Efi Koulouri wrote: > > Hi all, > > I am trying to implement an context-aware recommender in Mahout. As I > haven't use the library before I haven't a lot experience. So, I would > really appreciate your response! > > What I want to do is to implement the two context- aware approaches that > have been proposed, pre-filtering and post-filtering. The former filters > out the dataset based on the value of contextual factor before the > collaborative filtering while the latter rescores the recommendations after > the collaborative filtering. > > I have already read older similar questions regarding the context-aware > recommender implementation in mahout and I know that the post-filtering > method can be implemented using the IDRescorer. For the pre-filtering > approach there is the option to use the CandidateItemsStategy in case of > the item-based recommender. On the other hand if we want to implement this > approach using the user-bsed recommender no such option is available. > > In order to implement the pre-filtering using the user-based recommender, I > was thinking to filter out the unrelated user,items pairs from the dataset > before the creation of the data model. This means that the data model will > take as input a subset of the initial dataset. > Does this approach sound correct? There are some concerns regarding the > evaluation of the recommender. Does it have any impact on this? > > Thank you in advance! > > Regards, > Efi > >
Re: implementation of context-aware recommender in Mahout
The new cooccurrence recommender can use context as part of the user history, or as a method to bias or filter results. In any case you want to record all actions. Filtering results is easy and tossing all data but for one day of the week will reduce your cooccurrences and the quality of your data. 1) treat interaction with items at a day of week as a secondary action. The primary action is watch-video—it is what you want to recommend. Record video-view and _when_ they are viewed by each user. Create a cooccurrence indicator for video-view and a cross-cooccurrence indicator for viewed-on-day-of-week. 2) your query will be: history of video-views -> video-view indicators — this contribute in a non-contextual way history of viewed-on-day-of-week -> viewed-on-day-of-week cross-cooccurrence indicator — this contributes in a non-contextual way current time-of-day -> viewed-on-day-of-week boosted by some fairly high value OR use as a filter — this biases towards or filters b the current time-of-week. In this case you have 2 indicators, 1 of which is a time-of-day. You can use it as just another indicator (not as context) and/or as a boost or filter of results. The search engine supports all three methods in a single query if you have the indicator indexed. BTW this won’t work on the demo site since all you are doing is “Liking” which may corelated to the action of liking but not viewing of the video, which the demo site cannot record. On Mar 7, 2015, at 5:16 AM, Efi Koulouri wrote: Thanks for your reply! Actually, the context in pre-filtering serves as a query for selecting relevant data. An example of a contextual data filter for a movie recommender system would be: if a person wants to see a movie on Saturday, only the Saturday rating data is used to recommend movies. So,what I need for the rating prediction is the data relevent to the specific context. Regards, Efi On 7 March 2015 at 01:32, Pat Ferrel wrote: > The new Spark based recommender can easily handle context in many forms. > See the top references section here > http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html > > It does not use the IDRescorer approach at all so perhaps you should > describe what you want to use as context. > > In the demo site for the new stuff (a guide to online video) > https://guide.finderbots.com you’ll see a couple examples of “context”. > For instance you are viewing a video that has several genre tags. You’ll > see at least 3 lists of recommendations: > 1) people who like the video you are looking at also like these other > viedeos—non-personalized recs > 2) people who like this video liked these, from similar genres > 3) personalized recs from all genres based on your “liking” history > > Many other things can be used as context like time of day, location, > mobile or desktop, user profile attributes, etc. The way it does this is > through the search engine, which can take filters and boost certain item > attributes. So I could show only recommendations made in the same year as > the viewed movie or use the year to bias recommendations by boosting the > “release-date” field in the recommender query. The recommender is also > multimodal and so can use many user actions to better the quality of recs. > > Removing some of your data, in what you call pre-filtering may not get you > what you want. Removing data that is actual user behavior can reduce the > quality of recommendations so please give an example. > > On Mar 6, 2015, at 4:45 AM, Efi Koulouri wrote: > > Hi all, > > I am trying to implement an context-aware recommender in Mahout. As I > haven't use the library before I haven't a lot experience. So, I would > really appreciate your response! > > What I want to do is to implement the two context- aware approaches that > have been proposed, pre-filtering and post-filtering. The former filters > out the dataset based on the value of contextual factor before the > collaborative filtering while the latter rescores the recommendations after > the collaborative filtering. > > I have already read older similar questions regarding the context-aware > recommender implementation in mahout and I know that the post-filtering > method can be implemented using the IDRescorer. For the pre-filtering > approach there is the option to use the CandidateItemsStategy in case of > the item-based recommender. On the other hand if we want to implement this > approach using the user-bsed recommender no such option is available. > > In order to implement the pre-filtering using the user-based recommender, I > was thinking to filter out the unrelated user,items pairs from the dataset > before the creation of the data model. This means that the data model will > take as input a subset of the initial dataset. > Does this approach sound correct? There are some concerns regarding the > evaluation of the recommender. Does it have any impact on this? > > Thank you in advance! > > Regards
Re: implementation of context-aware recommender in Mahout
Ooops a several typos corrected below On Mar 7, 2015, at 7:05 AM, Pat Ferrel wrote: The new cooccurrence recommender can use context as part of the user history, or as a method to bias or filter results. In any case you want to record all actions. Filtering results is easy and tossing all data but for one day of the week will reduce your cooccurrences and the quality of your data. 1) treat interaction with items at a day of week as a secondary action. The primary action is video-view—it is what you want to recommend. Record video-view and _when_ they are viewed by each user. Create a cooccurrence indicator for video-view and a cross-cooccurrence indicator for day-of-week. 2) your query will be: history of video-views -> video-view indicators — this contribute in a non-contextual way history of day-of-week -> day-of-week cross-cooccurrence indicator — this contributes in a non-contextual way current day-of-week -> day-of-week boosted by some fairly high value OR use as a filter — this biases towards or filters by the current day-of-week. In this case you have 2 indicators, 1 of which is a day-of week. You can use it as just another indicator (non-contextual) and/or as a boost or filter results by it. The search engine supports all three methods in a single query if you have both indicators indexed. BTW this won’t work on the demo site since all you are doing is “Liking” and the day-of-week you liked something is not likely to correlated to “Liking” So I changed the example above to _viewing_ a video. Not all context will benefit your recommendations. But it is usually easy to experiment since the context applied at query time and as long at the data was used in training, no changes need to be made to the model (no pre-filtering) On Mar 7, 2015, at 5:16 AM, Efi Koulouri wrote: Thanks for your reply! Actually, the context in pre-filtering serves as a query for selecting relevant data. An example of a contextual data filter for a movie recommender system would be: if a person wants to see a movie on Saturday, only the Saturday rating data is used to recommend movies. So,what I need for the rating prediction is the data relevent to the specific context. Regards, Efi On 7 March 2015 at 01:32, Pat Ferrel wrote: > The new Spark based recommender can easily handle context in many forms. > See the top references section here > http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html > > It does not use the IDRescorer approach at all so perhaps you should > describe what you want to use as context. > > In the demo site for the new stuff (a guide to online video) > https://guide.finderbots.com you’ll see a couple examples of “context”. > For instance you are viewing a video that has several genre tags. You’ll > see at least 3 lists of recommendations: > 1) people who like the video you are looking at also like these other > viedeos—non-personalized recs > 2) people who like this video liked these, from similar genres > 3) personalized recs from all genres based on your “liking” history > > Many other things can be used as context like time of day, location, > mobile or desktop, user profile attributes, etc. The way it does this is > through the search engine, which can take filters and boost certain item > attributes. So I could show only recommendations made in the same year as > the viewed movie or use the year to bias recommendations by boosting the > “release-date” field in the recommender query. The recommender is also > multimodal and so can use many user actions to better the quality of recs. > > Removing some of your data, in what you call pre-filtering may not get you > what you want. Removing data that is actual user behavior can reduce the > quality of recommendations so please give an example. > > On Mar 6, 2015, at 4:45 AM, Efi Koulouri wrote: > > Hi all, > > I am trying to implement an context-aware recommender in Mahout. As I > haven't use the library before I haven't a lot experience. So, I would > really appreciate your response! > > What I want to do is to implement the two context- aware approaches that > have been proposed, pre-filtering and post-filtering. The former filters > out the dataset based on the value of contextual factor before the > collaborative filtering while the latter rescores the recommendations after > the collaborative filtering. > > I have already read older similar questions regarding the context-aware > recommender implementation in mahout and I know that the post-filtering > method can be implemented using the IDRescorer. For the pre-filtering > approach there is the option to use the CandidateItemsStategy in case of > the item-based recommender. On the other hand if we want to implement this > approach using the user-bsed recommender no such option is available. > > In order to implement the pre-filtering using the user-based recommender, I > was thinking to filter out the unrelated user,items pairs from the dat
Re: implementation of context-aware recommender in Mahout
The by far easiest way to build a recommender (especially for production) is to use the search engine approach (what Pat was recommending). Post filtering can be done using the search engine far more easily than using Java classes. On Sat, Mar 7, 2015 at 8:44 AM, Pat Ferrel wrote: > Ooops a several typos corrected below > > On Mar 7, 2015, at 7:05 AM, Pat Ferrel wrote: > > The new cooccurrence recommender can use context as part of the user > history, or as a method to bias or filter results. In any case you want to > record all actions. Filtering results is easy and tossing all data but for > one day of the week will reduce your cooccurrences and the quality of your > data. > > 1) treat interaction with items at a day of week as a secondary action. > The primary action is video-view—it is what you want to recommend. Record > video-view and _when_ they are viewed by each user. Create a cooccurrence > indicator for video-view and a cross-cooccurrence indicator for day-of-week. > 2) your query will be: > history of video-views -> video-view indicators — this contribute in a > non-contextual way > history of day-of-week -> day-of-week cross-cooccurrence indicator — this > contributes in a non-contextual way > current day-of-week -> day-of-week boosted by some fairly high value OR > use as a filter — this biases towards or filters by the current day-of-week. > > In this case you have 2 indicators, 1 of which is a day-of week. You can > use it as just another indicator (non-contextual) and/or as a boost or > filter results by it. The search engine supports all three methods in a > single query if you have both indicators indexed. > > > BTW this won’t work on the demo site since all you are doing is “Liking” > and the day-of-week you liked something is not likely to correlated to > “Liking” So I changed the example above to _viewing_ a video. Not all > context will benefit your recommendations. But it is usually easy to > experiment since the context applied at query time and as long at the data > was used in training, no changes need to be made to the model (no > pre-filtering) > > > On Mar 7, 2015, at 5:16 AM, Efi Koulouri wrote: > > Thanks for your reply! > > Actually, the context in pre-filtering serves as a query for selecting > relevant data. An example of a contextual data filter for a movie > recommender system would be: if a person wants to see a movie on Saturday, > only the Saturday rating data is used to recommend movies. > So,what I need for the rating prediction is the data relevent to the > specific context. > > Regards, > Efi > > On 7 March 2015 at 01:32, Pat Ferrel wrote: > > > The new Spark based recommender can easily handle context in many forms. > > See the top references section here > > http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html > > > > It does not use the IDRescorer approach at all so perhaps you should > > describe what you want to use as context. > > > > In the demo site for the new stuff (a guide to online video) > > https://guide.finderbots.com you’ll see a couple examples of “context”. > > For instance you are viewing a video that has several genre tags. You’ll > > see at least 3 lists of recommendations: > > 1) people who like the video you are looking at also like these other > > viedeos—non-personalized recs > > 2) people who like this video liked these, from similar genres > > 3) personalized recs from all genres based on your “liking” history > > > > Many other things can be used as context like time of day, location, > > mobile or desktop, user profile attributes, etc. The way it does this is > > through the search engine, which can take filters and boost certain item > > attributes. So I could show only recommendations made in the same year as > > the viewed movie or use the year to bias recommendations by boosting the > > “release-date” field in the recommender query. The recommender is also > > multimodal and so can use many user actions to better the quality of > recs. > > > > Removing some of your data, in what you call pre-filtering may not get > you > > what you want. Removing data that is actual user behavior can reduce the > > quality of recommendations so please give an example. > > > > On Mar 6, 2015, at 4:45 AM, Efi Koulouri wrote: > > > > Hi all, > > > > I am trying to implement an context-aware recommender in Mahout. As I > > haven't use the library before I haven't a lot experience. So, I would > > really appreciate your response! > > > > What I want to do is to implement the two context- aware approaches that > > have been proposed, pre-filtering and post-filtering. The former filters > > out the dataset based on the value of contextual factor before the > > collaborative filtering while the latter rescores the recommendations > after > > the collaborative filtering. > > > > I have already read older similar questions regarding the context-aware > > recommender implementation in mahout and I know that the post-filtering
Re: implementation of context-aware recommender in Mahout
Thanks for your help! Actually, I want to build a recommender for experimental purposes following the pre-filtering and post-filtering approaches that I described. I have already two datasets and I want to show the benefits of using a "context-aware" recommender. So,the recommender is going to work offline. I saw that the search engine approach is very interesting but in my case I think that building the recommender using the java classes is more appropriate as I need to use both approaches (post filtering,pre filtering). Am I right ? On 8 March 2015 at 16:08, Ted Dunning wrote: > The by far easiest way to build a recommender (especially for production) > is to use the search engine approach (what Pat was recommending). > > Post filtering can be done using the search engine far more easily than > using Java classes. > > > > On Sat, Mar 7, 2015 at 8:44 AM, Pat Ferrel wrote: > > > Ooops a several typos corrected below > > > > On Mar 7, 2015, at 7:05 AM, Pat Ferrel wrote: > > > > The new cooccurrence recommender can use context as part of the user > > history, or as a method to bias or filter results. In any case you want > to > > record all actions. Filtering results is easy and tossing all data but > for > > one day of the week will reduce your cooccurrences and the quality of > your > > data. > > > > 1) treat interaction with items at a day of week as a secondary action. > > The primary action is video-view—it is what you want to recommend. Record > > video-view and _when_ they are viewed by each user. Create a cooccurrence > > indicator for video-view and a cross-cooccurrence indicator for > day-of-week. > > 2) your query will be: > > history of video-views -> video-view indicators — this contribute in a > > non-contextual way > > history of day-of-week -> day-of-week cross-cooccurrence indicator — this > > contributes in a non-contextual way > > current day-of-week -> day-of-week boosted by some fairly high value OR > > use as a filter — this biases towards or filters by the current > day-of-week. > > > > In this case you have 2 indicators, 1 of which is a day-of week. You can > > use it as just another indicator (non-contextual) and/or as a boost or > > filter results by it. The search engine supports all three methods in a > > single query if you have both indicators indexed. > > > > > > BTW this won’t work on the demo site since all you are doing is “Liking” > > and the day-of-week you liked something is not likely to correlated to > > “Liking” So I changed the example above to _viewing_ a video. Not all > > context will benefit your recommendations. But it is usually easy to > > experiment since the context applied at query time and as long at the > data > > was used in training, no changes need to be made to the model (no > > pre-filtering) > > > > > > On Mar 7, 2015, at 5:16 AM, Efi Koulouri wrote: > > > > Thanks for your reply! > > > > Actually, the context in pre-filtering serves as a query for selecting > > relevant data. An example of a contextual data filter for a movie > > recommender system would be: if a person wants to see a movie on > Saturday, > > only the Saturday rating data is used to recommend movies. > > So,what I need for the rating prediction is the data relevent to the > > specific context. > > > > Regards, > > Efi > > > > On 7 March 2015 at 01:32, Pat Ferrel wrote: > > > > > The new Spark based recommender can easily handle context in many > forms. > > > See the top references section here > > > > http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html > > > > > > It does not use the IDRescorer approach at all so perhaps you should > > > describe what you want to use as context. > > > > > > In the demo site for the new stuff (a guide to online video) > > > https://guide.finderbots.com you’ll see a couple examples of > “context”. > > > For instance you are viewing a video that has several genre tags. > You’ll > > > see at least 3 lists of recommendations: > > > 1) people who like the video you are looking at also like these other > > > viedeos—non-personalized recs > > > 2) people who like this video liked these, from similar genres > > > 3) personalized recs from all genres based on your “liking” history > > > > > > Many other things can be used as context like time of day, location, > > > mobile or desktop, user profile attributes, etc. The way it does this > is > > > through the search engine, which can take filters and boost certain > item > > > attributes. So I could show only recommendations made in the same year > as > > > the viewed movie or use the year to bias recommendations by boosting > the > > > “release-date” field in the recommender query. The recommender is also > > > multimodal and so can use many user actions to better the quality of > > recs. > > > > > > Removing some of your data, in what you call pre-filtering may not get > > you > > > what you want. Removing data that is actual user behavior can reduce > the > > > quality of reco
Re: implementation of context-aware recommender in Mahout
Either architecture will work. Even if you want to pre-filter the data. The search engine can post-filter in the query. The pre-filter is to create a separate model for each day of the week, right? So works with any one. If you are relying on the evaluator implemented in Mahout then use the old java code because the new one doesn’t supply an evaluator. On Mar 8, 2015, at 10:21 AM, Efi Koulouri wrote: Thanks for your help! Actually, I want to build a recommender for experimental purposes following the pre-filtering and post-filtering approaches that I described. I have already two datasets and I want to show the benefits of using a "context-aware" recommender. So,the recommender is going to work offline. I saw that the search engine approach is very interesting but in my case I think that building the recommender using the java classes is more appropriate as I need to use both approaches (post filtering,pre filtering). Am I right ? On 8 March 2015 at 16:08, Ted Dunning wrote: > The by far easiest way to build a recommender (especially for production) > is to use the search engine approach (what Pat was recommending). > > Post filtering can be done using the search engine far more easily than > using Java classes. > > > > On Sat, Mar 7, 2015 at 8:44 AM, Pat Ferrel wrote: > >> Ooops a several typos corrected below >> >> On Mar 7, 2015, at 7:05 AM, Pat Ferrel wrote: >> >> The new cooccurrence recommender can use context as part of the user >> history, or as a method to bias or filter results. In any case you want > to >> record all actions. Filtering results is easy and tossing all data but > for >> one day of the week will reduce your cooccurrences and the quality of > your >> data. >> >> 1) treat interaction with items at a day of week as a secondary action. >> The primary action is video-view—it is what you want to recommend. Record >> video-view and _when_ they are viewed by each user. Create a cooccurrence >> indicator for video-view and a cross-cooccurrence indicator for > day-of-week. >> 2) your query will be: >> history of video-views -> video-view indicators — this contribute in a >> non-contextual way >> history of day-of-week -> day-of-week cross-cooccurrence indicator — this >> contributes in a non-contextual way >> current day-of-week -> day-of-week boosted by some fairly high value OR >> use as a filter — this biases towards or filters by the current > day-of-week. >> >> In this case you have 2 indicators, 1 of which is a day-of week. You can >> use it as just another indicator (non-contextual) and/or as a boost or >> filter results by it. The search engine supports all three methods in a >> single query if you have both indicators indexed. >> >> >> BTW this won’t work on the demo site since all you are doing is “Liking” >> and the day-of-week you liked something is not likely to correlated to >> “Liking” So I changed the example above to _viewing_ a video. Not all >> context will benefit your recommendations. But it is usually easy to >> experiment since the context applied at query time and as long at the > data >> was used in training, no changes need to be made to the model (no >> pre-filtering) >> >> >> On Mar 7, 2015, at 5:16 AM, Efi Koulouri wrote: >> >> Thanks for your reply! >> >> Actually, the context in pre-filtering serves as a query for selecting >> relevant data. An example of a contextual data filter for a movie >> recommender system would be: if a person wants to see a movie on > Saturday, >> only the Saturday rating data is used to recommend movies. >> So,what I need for the rating prediction is the data relevent to the >> specific context. >> >> Regards, >> Efi >> >> On 7 March 2015 at 01:32, Pat Ferrel wrote: >> >>> The new Spark based recommender can easily handle context in many > forms. >>> See the top references section here >>> > http://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html >>> >>> It does not use the IDRescorer approach at all so perhaps you should >>> describe what you want to use as context. >>> >>> In the demo site for the new stuff (a guide to online video) >>> https://guide.finderbots.com you’ll see a couple examples of > “context”. >>> For instance you are viewing a video that has several genre tags. > You’ll >>> see at least 3 lists of recommendations: >>> 1) people who like the video you are looking at also like these other >>> viedeos—non-personalized recs >>> 2) people who like this video liked these, from similar genres >>> 3) personalized recs from all genres based on your “liking” history >>> >>> Many other things can be used as context like time of day, location, >>> mobile or desktop, user profile attributes, etc. The way it does this > is >>> through the search engine, which can take filters and boost certain > item >>> attributes. So I could show only recommendations made in the same year > as >>> the viewed movie or use the year to bias recommendations by boosting > the >>> “release-dat
Re: implementation of context-aware recommender in Mahout
Efi, Only you can really tell which is best for your efforts. All the rest is our own partially informed opinions. Pre-filtering can often be accomplished in the search context by creating more than one indicator field and using different combinations of indicators for different tasks. For instance, you could create indicators for last one, two, three, five and seven days. Then when you query the engine, you can pick which indicators to try. That way the same search engine can embody multiple recommendation engines. I would also tend toward search-based approaches for your testing, if only because any deployed system is likely to use a search approach and thus testing that approach in your off-line testing gives you the most realistic results. On Sun, Mar 8, 2015 at 10:21 AM, Efi Koulouri wrote: > Thanks for your help! > > Actually, I want to build a recommender for experimental purposes following > the pre-filtering and post-filtering approaches that I described. I have > already two datasets and I want to show the benefits of using a > "context-aware" recommender. So,the recommender is going to work offline. > > I saw that the search engine approach is very interesting but in my case I > think that building the recommender using the java classes is more > appropriate as I need to use both approaches (post filtering,pre > filtering). Am I right ? > > On 8 March 2015 at 16:08, Ted Dunning wrote: > > > The by far easiest way to build a recommender (especially for production) > > is to use the search engine approach (what Pat was recommending). > > > > Post filtering can be done using the search engine far more easily than > > using Java classes. > >
Re: implementation of context-aware recommender in Mahout
Things got clearier with your help! Thank you very much On 9 March 2015 at 01:50, Ted Dunning wrote: > Efi, > > Only you can really tell which is best for your efforts. All the rest is > our own partially informed opinions. > > Pre-filtering can often be accomplished in the search context by creating > more than one indicator field and using different combinations of > indicators for different tasks. For instance, you could create indicators > for last one, two, three, five and seven days. Then when you query the > engine, you can pick which indicators to try. That way the same search > engine can embody multiple recommendation engines. > > I would also tend toward search-based approaches for your testing, if only > because any deployed system is likely to use a search approach and thus > testing that approach in your off-line testing gives you the most realistic > results. > > > On Sun, Mar 8, 2015 at 10:21 AM, Efi Koulouri > wrote: > > > Thanks for your help! > > > > Actually, I want to build a recommender for experimental purposes > following > > the pre-filtering and post-filtering approaches that I described. I have > > already two datasets and I want to show the benefits of using a > > "context-aware" recommender. So,the recommender is going to work offline. > > > > I saw that the search engine approach is very interesting but in my case > I > > think that building the recommender using the java classes is more > > appropriate as I need to use both approaches (post filtering,pre > > filtering). Am I right ? > > > > On 8 March 2015 at 16:08, Ted Dunning wrote: > > > > > The by far easiest way to build a recommender (especially for > production) > > > is to use the search engine approach (what Pat was recommending). > > > > > > Post filtering can be done using the search engine far more easily than > > > using Java classes. > > > >
Re: implementation of context-aware recommender in Mahout
Glad to help. You can help us by reporting your results when you get them. We look forward to that! On Tue, Mar 10, 2015 at 4:22 AM, Efi Koulouri wrote: > Things got clearier with your help! > > Thank you very much > > On 9 March 2015 at 01:50, Ted Dunning wrote: > > > Efi, > > > > Only you can really tell which is best for your efforts. All the rest is > > our own partially informed opinions. > > > > Pre-filtering can often be accomplished in the search context by creating > > more than one indicator field and using different combinations of > > indicators for different tasks. For instance, you could create > indicators > > for last one, two, three, five and seven days. Then when you query the > > engine, you can pick which indicators to try. That way the same search > > engine can embody multiple recommendation engines. > > > > I would also tend toward search-based approaches for your testing, if > only > > because any deployed system is likely to use a search approach and thus > > testing that approach in your off-line testing gives you the most > realistic > > results. > > > > > > On Sun, Mar 8, 2015 at 10:21 AM, Efi Koulouri > > wrote: > > > > > Thanks for your help! > > > > > > Actually, I want to build a recommender for experimental purposes > > following > > > the pre-filtering and post-filtering approaches that I described. I > have > > > already two datasets and I want to show the benefits of using a > > > "context-aware" recommender. So,the recommender is going to work > offline. > > > > > > I saw that the search engine approach is very interesting but in my > case > > I > > > think that building the recommender using the java classes is more > > > appropriate as I need to use both approaches (post filtering,pre > > > filtering). Am I right ? > > > > > > On 8 March 2015 at 16:08, Ted Dunning wrote: > > > > > > > The by far easiest way to build a recommender (especially for > > production) > > > > is to use the search engine approach (what Pat was recommending). > > > > > > > > Post filtering can be done using the search engine far more easily > than > > > > using Java classes. > > > > > > >