Natural language generation for sponsored-search advertisements

Bartz, Kevin; Barr, Cory; Aijaz, Adil

doi:10.1145/1386790.1386792

Cited by 17 publications

(9 citation statements)

References 16 publications

(11 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to attempt to capitalize the proper nouns in queries, we used a machine-learned system which searches for the query terms and examines how often they are capitalized in the search results, weighting each capitalization occurrence by various features (Bartz et al, 2008). Though the capitalization system provides 79.3% accuracy, using this system we see an only a small increase of accuracy in partof-speech tagging at 70.9%.…”

Section: Automatic Capitalizationmentioning

confidence: 99%

The linguistic structure of English web-search queries

Barr

Jones

Regelson³

2008

Proceedings of the Conference on Empirical Methods in Natural Language Processing - EMNLP '08

Self Cite

View full text Add to dashboard Cite

Web-search queries are known to be short, but little else is known about their structure. In this paper we investigate the applicability of part-of-speech tagging to typical Englishlanguage web search-engine queries and the potential value of these tags for improving search results. We begin by identifying a set of part-of-speech tags suitable for search queries and quantifying their occurrence. We find that proper-nouns constitute 40% of query terms, and proper nouns and nouns together constitute over 70% of query terms. We also show that the majority of queries are nounphrases, not unstructured collections of terms. We then use a set of queries manually labeled with these tags to train a Brill tagger and evaluate its performance. In addition, we investigate classification of search queries into grammatical classes based on the syntax of part-of-speech tag sequences. We also conduct preliminary investigative experiments into the practical applicability of leveraging query-trained part-of-speech taggers for information-retrieval tasks. In particular, we show that part-of-speech information can be a significant feature in machine-learned searchresult relevance. These experiments also include the potential use of the tagger in selecting words for omission or substitution in query reformulation, actions which can improve recall. We conclude that training a partof-speech tagger on labeled corpora of queries significantly outperforms taggers based on traditional corpora, and leveraging the unique linguistic structure of web-search queries can improve search experience.

show abstract

Section: Automatic Capitalizationmentioning

confidence: 99%

The linguistic structure of English web-search queries

Barr

Jones

Regelson³

2008

Proceedings of the Conference on Empirical Methods in Natural Language Processing - EMNLP '08

Self Cite

View full text Add to dashboard Cite

show abstract

“…By gathering all terms, we construct the extracted keywords vector. In order to boost trigrams first, bigrams second and unigrams third, we modify their relevance score with the following factor: boosted_score j = relevance_score j * k noOfGrams (4) where k is a free parameter (in our experiments we set it to (k = 100) and noOfGrams is the number of grams composing a term.…”

Section: Keyword Extraction Modulementioning

confidence: 99%

“…6, in the case of traffic maximization as the advertising goal, our two methods which use prediction, surpass the real results. In this experiment, the optimization process had started after the 4th week, because the advertiser until the 3rd week had been testing very few keyword options (3)(4) and the GA needs more testing data to perform a valid optimization. The important observation here compared with the stronger performance of the previous experiment was the use of much older and thus outdated data that did not correspond to valid receiving impressions and clicks in the ith week.…”

Section: Genetic Algorithm Performance On Optimizing Next Week's Perfmentioning

confidence: 99%

Toward an integrated framework for automated development and optimization of online advertising campaigns

Thomaidou

Liakopoulos

Vazirgiannis

2014

IDA

View full text Add to dashboard Cite

Creating and monitoring competitive and cost-effective pay-per-click advertisement campaigns through the websearch channel is a resource demanding task in terms of expertise and effort. Assisting or even automating the work of an advertising specialist will have an unrivaled commercial value. In this paper we propose a methodology, an architecture, and a fully functional framework for semi-and fully-automated creation, monitoring, and optimization of cost-efficient pay-per-click campaigns with budget constraints. The campaign creation module generates automatically keywords based on the content of the web page to be advertised extended with corresponding ad-texts. These keywords are used to create automatically the campaigns fully equipped with the appropriate values set. The campaigns are uploaded to the auctioneer platform and start running. The optimization module focuses on the learning process from existing campaign statistics and also from applied strategies of previous periods in order to invest optimally in the next period. The objective is to maximize the performance (i.e. clicks, actions) under the current budget constraint. The fully functional prototype is experimentally evaluated on real world Google AdWords campaigns and presents a promising behavior with regards to campaign performance statistics as it outperforms systematically the competing manually maintained campaigns.Google AdWords, Yahoo Search Marketing, and Microsoft AdCenter (the two latter have merged). 2 Today's most popular search-based advertising platform is Google AdWords having the largest share of revenues amongst its competitors. Google 3 2010 annual report showed that company's advertising revenues made up 97% of its revenues in 2008 and 2009, and 96% of its revenues in 2010. In 2010, Google advertising revenues (as an auctioneer) were $29 billion. 4 Search remains the largest online advertising revenue format, accounting for 46.5% of 2011 revenues, up from 44.8% in 2010. In 2011, Search revenues totaled $14.8 billion, up almost 27% from $11.7 billion in 2010. Search has remained the leading format since 2006, having strong sequential growth. 5 Effective keyword selection is one of the most important success factors for online advertising. Companies would like to advertise on the most effective keywords to attract only prospective customers and not uninterested browsing users. In addition, they need well-written ad creatives to attract more visitors and generate thus higher revenues.The major problem of an advertising campaign that takes into account only the suggestions of the most popular terms is that they are widely used, therefore the relevant keywords are quite competitive in terms of cost-per-click (CPC) cost. Another issue is that they are volume-based (i.e. number of monthly searches), which means these keywords will tend to drive more traffic to the campaign but not necessarily proportional conversions.Thus, to tackle the above issues we propose extracting keywords from a given landing page 6 and then extend them with f...

show abstract

“…For example, "coffeecolored women's t-shirt" is semantically/aesthetically better than "women's coffee-colored t-shirt". Knowing attributecluster labels facilitates obtaining an appropriate attribute order [108], [109]. The attribute-cluster scores help in determining the relevance of an attribute for the product category and hence, whether to include the attribute in the ad copy.…”

Section: Unsupervised Topic Discovery For Attribute Extractionmentioning

confidence: 99%

An overview of computational challenges in online advertising

Chatwin¹

2013

2013 American Control Conference

View full text Add to dashboard Cite

Online advertising is a large and rapidly growing business. The major players in the space, namely advertisers, publishers, and ad exchanges, are developing increasingly sophisticated systems, methods and tools to facilitate, manage, optimize and report on the performance of online advertising marketplaces and campaigns. Developing solutions that are both mathematically sound and practical draws on techniques from a variety of disciplines including machine learning, stochastic optimal control, information retrieval, data mining, natural language processing, and econometrics. In this paper, we provide an overview of the online advertising space, and identify, frame, and describe solution approaches to some of the major computational challenges in the space. We describe specific examples from industry applications, including ad inventory auctions, bidding and allocation strategies for ad inventory, inventory targeting, banner and landing page optimization, and performance estimation.

show abstract

Natural language generation for sponsored-search advertisements

Cited by 17 publications

References 16 publications

The linguistic structure of English web-search queries

The linguistic structure of English web-search queries

Toward an integrated framework for automated development and optimization of online advertising campaigns

An overview of computational challenges in online advertising

Contact Info

Product

Resources

About