Text Data Management and Analysis – Information Retrieval and Text Mining – Part II

The third and fourth sections of this book cover text data analysis and unified text data management analysis system.

Since some topics covered in the latter half of this book are new to me, two attributes of this book were particularly helpful and eased the learning curve: the pictorial illustrations and comprehensive references accompanying the text. The drawings used to illustrate concepts played an important role for me to grasp the gist fast. Due to the very broad coverage of topics by this book, it is impossible for the authors to elaborate on all the details in depth, particularly the areas are actively being researched. The many references included in these chapters direct me to the resources to follow up on, should I wish to dive further into a specific topic. I also like the discussion of evaluations included in nearly all chapters. We cannot improve what we cannot measure. For any system being implemented, designing appropriate evaluation methods guides the development of the rest of the system.

In the general framework of data mining, we humans can be viewed as subjective sensors, since the process of perceiving real-world events and describing or expressing thoughts about them in text data as a human being is very similar to how sensors like video cameras select and record some aspects of these same events. The data collected through these sensors, subjective or objective, can be in a text or non-text format. For the former, we can apply text mining; for the latter, more general data mining (e.g. image/video mining). Data mining software can be built as a combination of both or either. By applying data mining, a user hopes to derive actionable knowledge from the multi-modal and multi-source data. Through taking actions, the users in turn change the real world events and lead to newly updated data being collected. This process iterates as shown in this figure, cited from the course Text Mining and Analytics.

Section III Text data analysis starts with an entry-level introduction to text data analysis by discussing the potential applications, text vs non-text data and the landscape of text mining tasks. In the landscape of text mining tasks, the authors distinguish four types of task to infer knowledge from the data:

  • mining knowledge about the natural languages used to describe our observations, opinions, and so on.
  • mining the content of text data to gain knowledge about the observed world.
  • mining knowledge about the observers who produced the text data.
  • inferring other real-world properties and performing predictive analysis.

This section covers the following technical topics in details, and for each topic I provide a very brief description here. For further details, please refer to the book itself and the related research papers.

Word Association Mining

This entails discovering the two types of word relations: paradigmatic and syntagmatic. Words A and B have a paradigmatic relationship if we could substitute A for B or B for A without affecting the general understanding of the sentence where the word A or B occurs. This implies that A and B could be semantically or syntactically similar to each other, in other words, the two share a similar context. If words A and B have a syntagmatic relationship, they typically appear together in a context to express a coherent meaning.

The methods used for discovering these relations include examining the context of each word and compute certain similarity metrics of these contexts. To discover the correlated cooccurrences, we can use information based approaches, such as computing the conditional entropy of the occurrence of one word given the context of another word, or, computing the mutual information.

To evaluate the outcome, NDCG and average NDCG could be used to examine the ranked lists of relevance scores. Intrusion detection, using human judgement, can also check whether there is a word distinctively incoherent with the rest of the words in this word association group .

Text Clustering

Clustering allows us to discover the hidden structures in the data. The clustering techniques discussed in this chapter can be applied at both word and document level. Two categories of clustering technique are presented: similarity-based and model-based. Commonly, these clustering methods are unsupervised.

In a similarity based approach, defining one or a set of similarity metrics is a prerequisite. With a specified metric, clustering could be performed either top-down (i.e., divisive clustering) or bottom-up (i.e., agglomerative clustering). Typically in this scenario, the assignment of a word or document to a particular cluster is a hard binary one. A probabilistic model based approach typically allows soft assignment, meaning one data item could belong to multiple clusters, where each membership has a certain probability, and where all probabilities for this item sum up to one.

Text Categorization

Text Categorization goes one step further beyond clustering, as the goal here is to find the right category for text objects given a set of predefined categories. For example, we could design a program to automatically categorize each of my blog postings to the right topic category based on its content, computer science, business, history, fiction etc. What features would be useful to derive from the text for its categorisation? Recent studies show that combining the low-level lexical features with high-level syntactic features provide better performance in classification task than using either feature type alone. The book discusses three classification algorithms: k-nearest neighbors, naive Bayes, and linear classifiers.

Summarization

Text summarization builds on top of previous chapters and go one step higher up. The goal is to distill a large amount of text data into a concise summary. The key challenge here is to discover what are the important points conveyed in the full text. Two categories of methods are discussed in this chapter: extractive (selection-based) or abstractive summarization (generation-based).

The extractive approach typically splits the original document to sections, and selects relevant but not redundant sentences (or sub-sentences) from each section to form a summary without writing any new sentences. To achieve that, it applies discourse analysis and maximal marginal relevance.

The generation-based approach uses a language model to potentially write new sentences for the summary. The book gives an N-gram language model as an example, with n typically chosen to be about three to five. This approach may generate incomprehensible sentences due to its short-range nature. More advanced methods use other NLP techniques, such as named entity recognition and dependency parsers to extract the key entities and the relations among these entities from the text first. The authors refer them as actors and roles. The summary sentences are then generated by selecting from these identified entities and relationships in combination with a language model.

Topic Mining and Analysis

This chapter talks about using probabilistic topic models to discover latent topics in text data. The input to topic mining task is a set of documents and the number of topics k. The expected outputs are the k topics and, for each document, the probability that each topic is covered in the document, with the condition that the sum of the probabilities of all topic coverages for each document is one.

One approach is to use a mixture language model of two components: a background language model to account for the commonly occurring words in all documents and a topic specific model to represent the topic that we would like to discover. To discover more than one topic, this approach is generalised to a method called probabilistic latent semantic analysis (PLSA). The expectation-maximisation algorithm is used to estimate the PLSA.

Opinion Mining and Sentiment Analysis

As mentioned earlier, we human beings can be seen as subjective sensors of the real-world. The authors give the definition of “opinion” as cited from the course Text Mining and Analytics in the figure below.

To mine the users’ opinions and discover their sentiment, the book starts with three basic opinion representations: holder, target and content of the opinion; furthermore, it covers two enriched representation: opinion context and opinion sentiment.

The sentiment classification problem can be formulated as estimating a rating in the scale of 1 to n, given a document as input. One caveat is that these ratings are not independent. On the contrary, they reflect sentiment on some scale. As a result, we cannot treat it as a categorisation task of finding the most appropriate category for the document from a set of independent categories. We can adapt the binary logistic regression to multi-level for this task. However, a better approach with fewer parameters based on a similar idea is ordinal logistic regression. Furthermore, given reviews and their overall ratings as input, the latent aspect rating analysis discussed in the chapter can generate the major aspects covered in the reviews, the ratings on each aspect, and the relative weight placed on each aspect by each reviewer.

Joint Analysis of Text and Structured Data

This chapter discusses techniques for joint analysis of unstructured text data and structured data. One example is to use the non-text data as context to facilitate topic analysis of the text data. This is more formally known as contextual text mining. Three topics covered in this chapter are of great interest: contextual probabilistic latent semantic analysis, topic analysis with social networks as context and topic analysis with time series context.

The final section of the book talks about the text analysis operators, system architecture of a unified text analysis system, and the MeTA C++ data science toolkit provided by the research group of the authors.

   

Text Data Management and Analysis – Information Retrieval and Text Mining – Part I

This week, I return to my profession as a computer scientist and read the book titled “Text Data Management and Analysis – A Practical Introduction to Information Retrieval and Text mining” by ChengXiang Zhai and Sean Massung from University of Illinois at Urbana-Champaign. In Part I of the two-part blog post about this topic, I walk you through some key points of the first two sections of the book: Overview and Background, and Text Data Access. I leave the third section, Text Data Analysis, and the fourth section on a unified framework for text management and analysis to next blog post.

Overall, this book is very easy to follow. This might not be a very accurate projection from me who has worked on data-related topics in multiple areas of computer science over a decade. As far as computer science books are concerned, this statement stays true though. I would classify it as a textbook on information retrieval and text mining for 2nd or 3rd year undergraduates studying computer science, or, an entry-level book that opens the door to this field for people with a science background but specialised in other domains. If you prefer technical books of terse writing style, you may find yourself unsatisfied. It might seem to you that the authors did not make a great deal of effort to make the book concise. However, on the positive side, this means that there are very detailed explanations of concepts and how the algorithms and their associated maths formula are derived step-by-step. If you have not come across those before, you would appreciate this book’s thoroughness. There is a companion toolkit named the META toolkit available freely. It provides implementations of many techniques discussed in the book. Based on the material covered by this book, ChengXiang Zhai offers two courses on Coursera: Text Mining and Analytics and Text Retrieval and Search Engine.

Anyone who is reading this article would know that the amount of data produced per day has been increasingly dramatically over time. The characteristics of big data were summarized as 3-V: Volume (the quantity of data produced, collected, processed), Variety (incompatible data formats, non-aligned data structures, and inconsistent data semantics) and Velocity (the speed of data generation and subsequently speed requirement on analysis), by Doug Laney in his writing titled 3D Data Management. Later, the 3-V concept was expanded to 4-V (adding Veracity referring to the uncertainty of data) and 5-V (adding Value, referring to the ability to add value to business through insights derived from data analytics). Text data plays a significant role in this big data world. To process and exploit the ever-growing large amount of text data, there are two main types of services: text retrieval and text mining. The former is concerned with developing intelligent systems to help us to navigate the ocean of text data and access the most needed and relevant information efficiently and accurately. The latter focuses on discovering the purpose or intention of the text communication, deriving the semantic meaning, the underlying opinions and preferences of the users through the texts used, and by doing so assisting the users with decision making or other tasks. I am wary of using the word “knowledge” here, although it is standard practice in the writings of this field to see that extracted value from text as knowledge.

In this book, text information systems is described as offering three distinct and related functionalities: information access, text organisation and knowledge acquisition (text analysis). There are two typical ways of providing the access to relevant information to users: search engine and recommendation system. A search engine provides the users with relevant data, upon receiving certain queries from the users. Alternatively, it allows the users to browse the data through some hierarchical trees or other organisations, for example the browsing pane on Amazon site. This is typically referred to as pull model. It could be either personalised or not. A recommendation system takes a more active approach by pushing relevant information to a user as new information comes in with or without the updated user profile data. Hence it is referred to as push model. Text organisation is essential to make the information access and analytics effective. Although it is mostly hidden from a user’s perspective, it is this core part that glues the other parts of information system together. I include a drawing of a conceptual framework of text information systems from this book here for illustration purpose.

The prerequisite background knowledge for this domain include: probability and statistics, information theory and machine learning. Fear not though. Chapter 2 of the book discusses some of the basics. The appendix gives more detailed treatment on Bayesian statistics, expectation-maximisation, KL-divergence and Dirichlet prior smoothing.

The discussions on a few topics were interesting for me: statistical language models, the vector space and probabilistic retrieval models, all key components of search engine implementation (e.g., tokenizer, indexer, scorer/ranker, feedback schemes etc.), web indexing, link analysis, content-based recommendation and collaborative filtering. However, on the topics of link analysis and recommendation systems, I prefer Mining of Massive Datasets by Jure Leskovec, Anand Rajaraman and Jeff Ullman.

Happy reading!

     

Rubicon – the Triumph and Tragedy of the Roman Republic

My book of this week is Rubicon – the Triumph and Tragedy of the Roman Republic by Tom Holland. Rubicon tells the stories of the rise and fall of the Roman Republic, from the time when Lucius Tarquinius Superbus’ reign of Rome was demolished in a palace coup in 509 BC, the subsequent establishment of the Roman Republic, to the death of the first Roman Emperor, Augustus, in 14 AD.    

I was first introduced to Rubicon by Robert Harris, the author of a trilogy on the life of the great Roman orator Cicero. Over a month ago, while staying in London, I commented to one of my best friends that how marvellous it would be to meet and hear Robert Harris talk about the Roman history and its people. Unfortunately I have not had the opportunity of meeting him in person and firing up my numerous questions and fascinations to him about Cicero and the Roman Republic. However, BBC Radio 4 featured a book club event with Robert Harris shortly afterwards. Obviously I tuned in and listened to that episode a number of times. Robert Harris talked about why he chose to write about Cicero in this program. He credited that to the fact of reading an early manuscript of Tom Holland’s Rubicon. Before long, I found myself equipped with Rubicon’s ebook as primary reading format as it’s convenient to carry around, hardcopy for flipping through and cross referencing, and finally audiobook for listening while running or driving. There are a noticeable amount of words and sentences that are different in the audiobook than those in the ebook, although not affecting the readability much nor leading to gross misunderstanding.

Rubicon is the most challenging book for me so far this year. Although, as a child, I read Chinese history and literature extensively, and later various genres of books in English (mostly literature, biographies, science, engineering and history books), fundamentally I am a computer scientist. Rubicon is the first time that I read a full volume of the history of the Roman Republic. Unsurprisingly, the writing of this book is very different from the trilogy on Cicero I read and wrote about earlier this year, as it not only has a captivating storytelling part, but also probably half of it contains discussions and analysis of the political and social struggles of the people in Rome, its provinces and far beyond. That said, a number of the metaphors that we use nowadays can be traced back to the Roman time over two thousand years ago, for example, the term “Cross the Rubicon” with the meaning that once we pass a certain point there would be no return, similarly the phrase “the die is cast”, might have both originated from the event that Julius Caesar led his army crossing the Rubicon River in 49 BC as an act of insurrection and treason. Discoveries like these add extra pleasure to my experience of reading history books like Rubicon.

In this book, to me, the paradoxical features of the Roman Republic are most distinctive. This is reflected in multiple facets of Roman society, for example, the wide division of two classes and their mutually shared devotion to the community. In the Roman Republic there is nothing resembling a middle class. Everybody is either plebeian or patrician. “The central paradox of Roman society – that savage divisions of class could coexist with an almost religious sense of community – had evolved through the course of its history. A revolution against the extractions of authority had, of course, inspired the Republic’s very foundation. Even so, following the expulsion of Tarquin and the monarchy, the plebeians had found themselves just as tyrannized by the ancient aristocracy of Rome, the patricians, as they had ever been by the kings……Indeed, in the early years of the Republic’s history, Roman society had come perilously close to ossifying altogether. The plebeians, however, refusing to accept that they belong to an inferior caste, had fought back in the only way they could – by going on strike……Here they would periodically threaten to fulfil Remus’s original ambitions by founding an entirely new city. The patricians, left to stew in their own hauteur across the valley, would gracelessly grant a few concessions. Gradually, over the years, the class system had become ever more permeable. The old rigid polarization between patrician and plebeian had begun to crack.” This should sound very familiar to most people who read world history and follow a little social and political movements of the modern world.

More on the paradox of Roman society: “The privileges of birth, then guaranteed nothing in Rome. The fact that the descendents of a goddess might find themselves living in a red-light district ensured that it was not only the very poor who dreaded the consequences of failure. At every social level the life of a citizen was a grueling struggle to emulate – and, if possible, surpass – the achievements of his ancestors. In practice as well as principle the Republic was savagely meritocratic. Indeed, this, to the Romans, was what liberty meant. It appeared self-evident to them that the entire course of their history had been an evolution away from slavery, toward a freedom based on the dynamics of perpetual competition. The proof of the superiority of this model of society lay in its trouncing of every conceivable alternative. The Romans knew that had they remained the slaves of a monarch, or of a self-perpetuating clique of aristocrats, they would never have succeeded in conquering the world. “It is almost beyond belief how great the Republic’s achievements were once the people had gained their liberty, such was the longing for glory which it lit in every man’s heart.”……For all the ruthlessness of competition in the Republic, it was structured by rules as complex and fluid as they were inviolable. To master them was a lifetime’s work. As well as talent and application, this required contacts, money and free time. The consequence was yet further paradox: meritocracy, real and relentless as it was, nevertheless served to perpetuate a society in which only the rich could afford to devote themselves to a political career. Individuals might rise to greatness, ancient families might decline, yet through it all the faith in hierarchy endured unchanging.”

The women written about in this book are fascinating characters to me: the Sibyl and her prophecies, the Egyptian Queen Cleopatra, the notoriously unfaithful and manipulative Clodia whose social standing was destroyed mercilessly by Cicero during the trial of Caelius and who subsequently vanished from the public eye, Fulvia whose political involvement was not to be underestimated although largely hidden behind the men she supported, and a few others. Cleopatra was written about in great depth in this book for her close association with Mark Antony and Julius Caesar. I am not a historian on any of these female characters. Nevertheless, I wonder whether what we now know of them are twisted facts mixed with some fanciful and maybe even false projections from what could be found in the broken historic records. Those records were written in an era during which societal judgement on a woman was archaic and misogynistic. Here is a passage from the book on Clodia: “For any woman, even one of Clodia’s rank, dabbling in politics was a high-wire act. Roman morality did not look kindly on female forwardness. Frigidity was the ultimate marital ideal. It was taken for granted, for instance, that a matron has no need of lascivious squirmings – anything more than a rigid, dignified immobility was regarded as the mark of a prostitute. Likewise, a woman whose conversation was witty and free laid herself open to an identical charge. If she then compounded her offences by engaging in political intrigue, she could hardly be regarded as anything other than a monster of depravity.” There is also a short piece about Aurelia Cotta, Julius Caesar’s mother, widely praised and respected by Roman people for breastfeeding her children. One wonders why it is even anybody else’s business to have an opinion on whether a mother decides to breastfeed or not.

From reading the trilogy of Cicero, I learned that Cicero admired Cato greatly for his unyielding character. In Rubicon, Tom Holland went into more depth of portraying Cato. An example passage is included here. Throughout this book, Cato stood out as the character with more integrity and principle than any others. “Marcus Porcius Cato had a voice that boomed out across the Senate House floor. Rough and unadorned, it appeared to sound directly from the rugged, virtuous days of the earlier Republic. As an officier, Cate had ‘shared in everything he ordered his man to do. He wore what they wore, ate what they ate, marched as they marched.’ As a civilian, he made a fashion out of despising fashion, wearing black because the party set all sported purple, walking everywhere, whether in blazing sunshine or icy rain, despising every form of luxury, sometimes not even bothering to put on his shoes. If there was more than a hint of affectation about this, then it was also the expression of a profoundly held moral purpose, an incorruptibility and inner strength that the Romans still longed to identify with themselves, but had rather assumed were confined to the history books. To Cato, however, the inheritance of the past was something infinitely sacred. Duty and service to his fellow citizens were all. Only after he had fully studied the responsibilities of the quaestorship had he been prepared to put himself up for election. Once in office, such as his probity and diligence that it was said he ‘made the quaestorship as worthy of honor as a consulship’. Plagued by a sense of its own corruption as it was, the Senate was not yet so degenerate that it could fail to be impressed by such a man.”

Many other significant characters are covered in the book along the historic river flowing through the formation of the Roman Republic, its expansion and conquering of many territories, many tortuous turns of its fortune, and finally its fall. A few more examples of very luminous characters are Lucius Cornelius Sulla Felix, Pompey the Great, Marcus Licinius Crassus, Marcus Antonius, Julius Caesar and Augustus. I hope to come back to write an addition on these characters, especially Augustus, in the near future.

Reading Rubicon and writing a summary about it is the toughest test since I started my one-book-a-week project this year. Outside work, I lost my opinions on nearly all basic daily activities, for example, what to eat or drink. Who would care about these trivialities, if you are immersed in this glorious, heart-wrenching, treacherous Roman history? If not for the fact of hosting an alumni event later this Sunday, I should be living in the Roman Republic, walking around the Palatine and Aventine hills for a little bit longer before the end of this weekend.

My Life in Advertising

My book of the week is an autobiography by Claude C. Hopkins, My Life in Advertising, written in 1927.

Walter Weir once praised this book: There are few pages in My Life in Advertising which do not repay careful study – and which do not merit rereading. Before your eyes, a successful advertising life is lived – with all that went to make it successful. The lessons taught are taught exactly as they were learned. They are dished up dripping with life. It is not a book, it is an experience – and experience has always been the great teacher.

When I was reading some books about entrepreneurship, a recurring message seems to be that many startup founders are excellent at exercising their visions and building great products, but are not good at selling these products. I pondered about this for a while. Multiple questions came to my mind:

  • Why is this the case?
  • What can be done about it?
  • What skills are required for salesmanship?
  • Could we learn it if we do not have that natural talent?
  • If so, how to learn and practice?

This led me to search for a book on salesmanship and I came across this book by Claude C. Hopkins. Among many titles, I chose this one for two reasons. First, I hoped a retrospective reflection of one’s life time work on advertising would be more interesting and have greater depth than books that focus on teaching methods rather than principles and likely filled with buzzwords. Second, I was curious about how advertising was done in the beginning of 20th century with very different media channels from those of today and whether the practices and principles learned back then would still be valuable for us.

This book certainly did not disappoint me. It not only served well in terms of answering my questions and opening up my mind to advertising, but also offered me a rich experience of going through many advertising missions together with the author such that the principles and insights presented in the book come naturally to me. Furthermore, through those real-world examples, I, as a reader, am free to draw my own conclusion and form thoughts other than those the author presented. I share with you here a few thoughts derived from the book that resonate with me most.

In my view, Hopkins credited much of his success in advertising to poverty. Poverty led Claude to live among the common people, to know them, to understand their wants and impulses, their struggles and economies, their simplicity. His early years were full of hardship. Subsequently the very down-to-earth style that he developed offered him a window into ordinary people’s lives and to stay connected with them. Here in Claude’s own words: I am sure I would fail if I tried to advertise the Rolls-Royce, Tiffany&Co, or Steinway pianos. I do not know the reactions of the rich. But I do know the common people. I love to talk to laboring-men, to study housewives who must count their pennies, to gain the confidence and learn the ambitions of the poor boys and girls. Give me something which they want and I will strike the responsive chord. My words will be simple, my sentences short. Scholars may ridicule my style. The rich and vain may laugh at the factors which I feature. But in millions of humble homes the common people will read and buy. They will feel that the writer knows them. And they, in advertising, form 95 percent of our customers.

Hardworking is an attribute shared among people who are successful in their endeavors. But, how many could match Claude’s level of industry? “I have supported myself since the age of nine. Other boys, when they went to school as I did, counted their school work a day. It was an incident to me. Before school, I opened two school houses, built the fires, and dusted the seats. After school I swept those school houses. Then I distributed the Detroit Evening News to sixty-five homes before supper. On Saturdays I scrubbed the two school houses and distributed bills. On Sundays I was a church janitor, which kept me occupied from early morning until ten at night. In vacations I went to the farm, where the working time was sixteen hours a day. When the doctor pronounced me too sickly for school, I went to the cedar swamp. There work started at 4:30 in the morning….Yet it never occurred to me that I was working hard. In after years I did the same in business. I had no working hours. When I ceased before midnight, that was a holiday for me. I often left my office at two in the morning. Sundays were my best working days, because there were no interruptions. For sixteen years after entering business I rarely had an evening or a Sunday not occupied by work….The man who works twice as long as his fellows is bound to go twice as far, especially in advertising…There is some difference in brains, of course, but it is not so important as the difference in industry. The man who does two or three times the work of another learns two or three times as much. He makes more mistakes and more successes, and he learns from both. If I have gone higher than others in advertising, or done more, the fact is not due to exceptional ability, but to exceptional hours….Frugality and caution kept me from disaster, but industry taught me advertising and made me what I am.”

Industry alone is not sufficient for great success, if one does not love his work while ploughing the field of his choice. “What others call work I call play, and vice versa. We do best what we like best.

On distributing credit and being fair, there are a few stories told in this book and a few lessons that we could draw from that. One key message that echoes what the lecturer of my Stanford leadership class repeated: I for responsibility, we for credit. One example passage from the book: About the only disagreements I had with Mr. Lasker referred to his desire to overpay me. That attitude I consider a vital factor in success. An absolutely fair division. One on the crest of the wave may over-play his hand for a little time, but not for long. Business is money-making, and associates will find a way to eliminate anyone who claims too large a share.

One great lesson about advertising is to start small, test and gather data, learn from the data, improve and iterate. This book was written in 1920s about salesmanship, but the concept presented here is the same as lean startup and agile development. This reaffirms my view that reading or getting to know the fields other than the one we mostly practice in can tremendously broaden our view and improve how we practice in our own field. On the importance of experimenting, Hopkins wrote “but we find that some methods which succeed in one line cannot be applied to another. We find that some methods which are profitable are not one-fourth so effective as others. So, regardless of principles, we must always experiment.

There are many great principles that Hopkins summarised from his decades of experience in advertising. I can not enumerate all here, but include a few that were most refreshing for me when I read them the first time.

  • Brilliant writing has no place in advertising. A unique style takes attention from the subject. Any apparent effort to sell creates corresponding resistance. Persuasive ability arouses the fear of over-influence.
  • Never try to show off. You are selling your product, not yourself. Do nothing to cloud your objective. Use the shortest words possible. Let every phrase ring with sincerity.
  • Aim to get action.
  • Ads should tell the full story. People do not read ads in series. The advertiser who today attracts them may not again get attention for months. So, when you get a reading, present all your arguments. In an advertising campaign, we find facts which appeal, and we retain them. We find facts which don’t appeal, and we drop them. We find these things out by featuring our various claims in headlines. We find that one lead brings a great deal of interest, while another brings little or none. So we gauge our appeals accordingly.
  • When we say such things as, “The best product in existence,” “The supreme creation of its kind,” we may arouse only a smile at our frailties. No resentment may be engendered. But whatever else we say is discounted…..On the other hand, when you state actual figures, definite facts, they accept them at par.
  • All experience in advertising proves that people will do little to prevent troubles. They do not cross bridges in advance. They will do anything to cure troubles which exist, but legitimate advertising has little scope there. All are seeking advantages, improvements, new ways to satisfy desires.

I applied what I learned by reading this book on two occasions this week. One was to make a maiden “sales pitch” about the roses I grow and my floral arrangements. The other was a discussion about acquiring a large customer base for a product currently under development. Based on the feedback received, both have gone well, with lots of space for further improvement on my part, which I always expect to be the case.