Marketing in the business world.

IntroductionWith competition so intense, many businesses is forced to find innovative ways to increase the quality and sales of their products, from manufacturing to inventory to marketing, using advance technology that is now readily available for anyone willing to adopt it. This is exactly why over the last few years’ information technology has taken a firm grasp and continues to gain momentum. The business world is competitive and everyday presents new and difficult challenges.   Companies must rely on the most effective marketing and sales strategy in order to remain in front of their competition.   In the business world, most companies nowadays use the web technology and Internet to ensure these goals are both surpassed and met. This paper discusses in detail the approach of companies to marketing in the business world.Marketing in the Business WorldIn the business world, effective marketing becomes increasingly tied up with the Internet and other electronic media, ma king the most of the Internet and other new technologies is important to a business success — from the brand image portrayed on its Web site to the development, maintenance and enhancement of customer relationships. In these increasingly uncertain and cynical times, marketing in the business world unpicks the challenges of e-marketing for many types of business.The Web is very information concentrated environment. Extensive amounts of information can be integrated, collected, presented, processed, and accessed through the Web by both consumers and marketers. Marketers now can track comprehensive information for all consumer interactions, not just select examples. On the other hand, soft market qualities, such as reliability and reputation, maybe more difficult to evaluate and gather.Consumers can increase more market information for criteria evaluation; however possible problems of information excess may increase recognised searching costs (Head et al. 2000). Marketers may pr ovide tools to facilitate consumer information collection, but may have to restructure their marketing strategies since competitors can also gather market information and match price differences.Peterson et al. (1997) argues that the Web will have major effect on communication, will influence transactions, but will have no effect on distribution unless the good is based on digital assets. Businesses, which consider the Web as a marketing instrument, primarily utilise it as a communication means to take advantage of its benefits in lower costs, personalisation, interactivity, digitisation, automation, and constant communication. Most businesses who do not currently sell their products/services through their web sites hold this view. Many businesses create their site to foster better communication and public relations. Consumers can obtain corporate information, and may be encouraged to subscribe to customised electronic flyers/newsletters.Mahajan and Wind (1989) explained that Web is a market discontinuity. Companies, which deal with the Web channel as a promising new marketplace that helps a complete range of interaction, may find that it co competes and -exists with traditional markets. New importance must be identified and valued by consumers to compete and participate with existing market channels. For instance, customers must value the ease and flexibility of using Web. Companies need to deal with market positions/boundaries to fit the new value with customers' needs, and look for proper business models in order to achieve these needs. Customer base will be new and existing groups with unique value principle, which may be difficult to attain through traditional channels.In addition to using the Web as a marketing tool and for information searching, consumers can increase their market power through high involvement in business processes and virtual communities. These are new Web-centered strategies, which are restricted in traditional markets by fixed physi cal assets and slow transfer of information (Werbach 2000). Some researchers have noted the commercial effects of Web groups (Kozinets 1999). Communication convenience allows consumers to form communities outside their traditional work or family groups.Such communities accumulate information or knowledge by learning from the experiences of individual community members. Information collection costs and times are often reduced for individuals within a community. Opinions from other consumers are often more valued than messages from marketers, especially in a Web market, where trust is critical and more difficult to build. Not only do groups have a strong effect on purchasers' decision-making, but they may also affect market variables such as products and prices.For example, stocks that capture the attention of participants in chat rooms can move noticeably in price (Bruce 1997). The Web, as an interactive marketplace, also gives the consumer data selection and personalisation power. C ustomers can select information of interest and personalise presentation forms for their own use. Personalized Web pages, which can be constructed fairly easily, increase customer power. Customers on the Web have greater control over what they view and examine. They can select their own path through the information network, process the data, or initiate communication with marketers.There is a general consensus that the industrial organisational impacts of using e-commerce as marketing tool will reflect two developments: 1) the expansion of relevant geographical markets, and 2) increased competition in those markets. The two changes are related (Globerman et al. 2001). However, E-commerce businesses are characterised by high market capitalisations, which are reflected in the perception of their business models by investors (Venkatraman 2000). Primarily, businesses in the Internet carry a 30% marketing budget in order to reach more customers.Specifically, as electronic commerce makes it less costly to identify beneficial transactions across a wider range of potential transactors, it should lead to an increased integration of markets that are currently segmented by high transactions costs across geographical space. In addition, geographically larger markets are ordinarily more contestable than smaller markets. In this perspective, one well-known international business professional stated that electronic commerce implies the end of borders and geography as industrial organisation constructs (Kobrin, 1995).Decision to include international buyers as part of the customer base for a business's Web site, changes must be made to the site in order to promote global consumers to buy products. Internationalising the Web site will aid in the attraction and retention of foreign users by allowing them easier access to the information and functions it presents in a standardised, more simplified manner.Most businesses lack the expertise and resources to create separate Web sit es along with the independent channels of marketing, distribution and production facilities needed for each target market they intend to enter. As a result, these businesses are expected to enter into the business world on a smaller scale, sticking to stages one and two when redesigning their Web sites with a global focus.ConclusionThe Internet has made an outstanding impact within its first decade of business use. Marketing professionals have been quick to realise the opportunities provided by the Internet. Particularly, they recognised that business could be improved by integrating the Internet with direct marketing practices such as database marketing. The challenge for information systems practitioners and professionals is to understand these opportunities with the accessible technologies in the relative constraints of an organisation.It is included that e-commerce which is driven by the exponential growth of the Web is the most common marketing practices taking advantage of the Web by utilising information technology to sell large quantities of products and to become more responsive to the individual. The role of the Internet, in particular, of the electronic commerce web sites, has been recognised as a marketing tool for attracting and maintaining customers.ReferenceBruce, C. (1997). Welcome to my parlor. Marketing Management, 5 (4), 11-24.Globerman, S., Roehl, T. and Standfird. (2001) Globalization and Electronic Commerce: Inferences from Retail Brokering. Journal of International Business Studies. Volume: 32. Issue: 4.Head, M., Archer, N.P., & Yuan, Y. (2000). World wide web navigation aid. International Journal of Human-Computer Studies, 53 (2), 301-330.Kobrin, S.J. 1995. Regional Integration in a Globally Networked Economy. Transnational Corporations, 4 (2): 15-33.Kozinets, R.V. (1999). E-tribalized marketing?: The strategic implications of virtual communities of consumption. European Management Journal, 17 (3), 252-264.Mahajan, V., & Wind, J. (1989) . Market discontinuities and strategic planning: A research agenda. Technological Forecasting and Social Change. 36 (August), 185-199.Peterson, R., Balasubramanian, S., & Bronnenberg, B.J. (1997). Exploring the implications of the internet for consumer marketing. Journal of the Academy of Marketing Science, 25 (4), 329-346.Venkatraman, N. (2000) Five steps to a dot.com strategy: How to find your footing on the Web, Sloan Management Review, 41(3), 15-28.Werbach, K. (2000). Syndication: The emerging model for business in the Internet era. Harvard Business Review 78 (3), 85-93.

How the theme of servant and master in the Tempest is connected to the Essay

How the theme of servant and master in the Tempest is connected to the scenic progression - Essay Example ..' however further on lines 348 and 349 we learn that Caliban tried to rape Miranda and was not sorry for it. He wished he could have succeeded as he says, 'Thou didst prevent me - I had peopled else/This isle with Calibans.' This seems to be the reason for Prospero treating Caliban this way and shows Calibans savage, vindictive nature. This quote shows they have exploited him by making him work for them and treating him as an inferior by relating him to an animal by calling him names such as 'tortoise'. However Prospero also says, 'we cannot miss him' which shows how Prospero and Miranda rely on him and should treat him with a little bit more respect. Other ways in which we see Prospero as the colonizer are he forced Caliban to learn his language and Miranda calls Calibans language 'gabble'. He also insults Calibans mother and refers to Calibans race as 'vile'. This is how colonizers would have behaved. They would see their race as far superior. Prospero is also viewed as a colonizer because he uses force to control Caliban. He says, '...I'll rack thee with old cramps/Fill all thy bones with aches, make thee roar...' This shows that if Caliban is not willing to obey and do things for Prospero he will make him do it by inflicting pain upon him. This is how some f the colonizers would have behaved because they wanted control. This shows Caliban is paying fo... This is how some f the colonizers would have behaved because they wanted control. We see Caliban as the colonized native because he was stupid enough to show Prospero all the resources f the island and then he lost his island to Prospero. Caliban tells us this by saying, '...And then I loved thee And showed thee all the qualities o' the 'sle, The fresh springs, brine-pits, barren place... Cursed be I that did so!...' This shows Caliban is paying for his actions and wishes he had never have shown Prospero the good qualities f the island. This mirrors how the natives behaved. They genuinely welcomed the colonizers and treated them well. Again in Act 2 Scene 2 Caliban the same mistake again when he says to Stephano, ' I'll show thee every fertile inch o' th'island. And I will/kiss thy foot - I prithee be my god.' Here we can see that Caliban is making the same mistake again which leads us to believe he really is dumb and fits into the category f the stereotypical native. However here Caliban can be seen as quite intelligent as he sees Stephano as a softer master and an easier way to eventually be free. We know this because Caliban sings, 'No more dams I'll make for fish, Nor fetch in firing At requiring, Nor scrape trenchering, nor wash dish, ....Has a new master-get a new man. Freedom, high-day, high-day freedom' Although Caliban appears to be smart here he says in his song he will no longer have to do these tough tasks Prospero makes him do even though he promised to do them for Stephano. When Prospero first came to the island he was nice to Caliban they way Stephano is being but that could change when Caliban does not do what he promised and Stephano could end up being a harder master. Evidence that Caliban was colonised by Prospero is that he is

Manipulating Data Essay Example | Topics and Well Written Essays - 500 words

Manipulating Data - Essay Example It is the result of an old code that is modified over the years plenty of times. Another aspect is that changing one part of the code would have unpredictable effects on all the other parts of the program, just as a bowl of spaghetti where pulling one can affect all the other strands. Thus the complex structure is named after spaghetti. Spaghetti code is caused mainly by inexperienced programmers following their mandates and creating a complex program which is being modified by several other people previously. Structured programming however decreases the chance of spaghetti code (Dixit, 2007, p. 92). The structured programming was a method formed in 1966 as a logical programming method which is a precursor to the object-oriented programming. This programming method is aimed to improve the quality, clarity, and development time of computer programs through the extensive use of block structures and subroutines instead of simple tests such as GOTO statements resulting in spaghetti code which makes it difficult to maintain and follow (Agarwal, 2009, p. 253). Modular programming has been functioning since the 1970s as a technique which subdivides a computer program in various other sub-programs. It separates the computer programs into individual and independent modules. It is a separate software component which is used with many other applications and functions in the system. The functions which are similar are grouped together while the separate functions are grouped as separate units. Object-oriented programming can be used with modular programming as it allows multiple programmers to work on divided programs independently (Mitchell, 2003, p. 239). Object-oriented programming is the method which is most commonly used today. It provides a programming model based on objects as it integrates the code and data by using objects. An object can be the abstract data type which has a state and behavior both. These objects can also be like real

Risk Assessment Essay Example | Topics and Well Written Essays - 250 words

Risk Assessment - Essay Example Pesticides sprayed on plants are not just sprayed on plants, but soak into the soil on which we walk and get into the underground water supply, contaminating it so that it becomes unsafe to drink. I also do not want to put anything into my body that I cannot read on a label and pronounce and understand, and speaking from the heart, I cannot understand the name of any pesticide, except to tell you that anything ending in â€Å"zine† is bad. Pesticides and herbicides should be tested for their potential to harm humans by putting them into a controlled sample of water and observing changes over time. If the water were to turn an unnatural color, such as brown or gray, the pesticide would then be deemed unsafe. It should also be tested on samples of soil by placing the controlled pesticide sample into the soil and monitoring plant growth throughout the sample. For further testing, we should introduce fish into a larger, yet still controlled sample of water, and if the pesticide sample introduced to that environment caused any harm to the fish, then we would know it would be unsafe for humans. As a consumer, we have a responsibility to not only read labels on the food that we purchase, but to know where our food comes from. Simply choosing to buy organic is not enough. We must know, at all times, if the can of corn or the bag of onions that we are about to pick up in the grocery store came from a place where chemicals are used on the food. This is our responsibility as a consumer, to make sure that we are aware of what is going into our

Social Responsibility and Financial Performance Literature review

Social Responsibility and Financial Performance - Literature review Example In order to conduct a literature review, two main themes are being highlighted including the relationship of CSR and financial performance and the relationship of CSR and Shareholders Value (Aras et al., 2010). A literature review is carried out by incorporating conceptual and theoretical frameworks (Aupperle et al., 1985). The theoretical framework provides relevant theories and models associated with corporate social responsibility and its impact on financial performances and shareholders’ value. The conceptual framework would improvise the need for research on the basis of previous researchers by various scholars and be examining their objectives, the methodology adopted and research findings. This provides insight on how different variables identified in themes chosen are linked together and are a significant part of in the corporate sector. The emergence of corporate scandals, financial responsibilities, and environmental threats has sensitized the society to be concerned with the social and ethically responsible behavior of companies. The concept of Social Responsibility (CSR) refers to the role of businesses towards the society. Cane and Matten (2010) in their study highlights, ‘it is by now fairly widely accepted that businesses do indeed have responsibilities beyond simply making a profit’ (Crane & Matten, 2010). A number of concepts such as Corporate Citizenship, Corporate Sustainability, and Corporate Social Responsibility are listed with regard to social responsibilities.  

Contemporary Issues in Financial Services Essay

Contemporary Issues in Financial Services - Essay Example These factors can be consumed in the present time or utilized with an aim of producing in the future. Economic activity is taken as real because real resources are utilized to come up with new products that can be bought and used. On the other hand, the financial system mainly focuses on either circulating funds around to enable people buy goods and services or assist people exchange ownership of resources vital for production purposes. It is worth mentioning that activities of the real economy are vital for human survival. The real economy is responsible for the production of consumer goods, food, lighting, heating, and entertainment to mention but a few (Mankiw 2008, p.162). The role of the financial system is to facilitate the production of such goods by ensuring that funds are available when needed. Therefore, the structure of the financial system comes to the fore since it would offer alternative financing windows. These institutions provide an avenue through which large or smal l manufacturing sectors can access finance, which would boost production. The role undertaken by the financial system in coordinating economic activities, which includes the cost of finance, profitable investment outlets, and availability of investment funds, has a direct relationship with the structure of the financial system. ... However, it appears that most economists are in agreement that a link exists between the financial system and the real economy (Burton & Brown 2009, p317). The structure and role of the financial system The financial system is made of financial institutions and financial markets. Examples of the financial institutions include the banking system (banking institutions and the central bank), and non-bank financial institutions (insurance and pension funds). The financial markets are made of foreign exchange markets and money markets, capital markets, which include bond markets, private debt securities, public debt securities, and equity markets, and the derivatives markets. The current literature work has not fully addressed the issue of the role and structure of the financial system essential in facilitating and sustaining growth. There are two general agreements in relation to the role and structure of financial systems; first, countries with good working banks and financial markets c an increase their growth pace. Secondly, better functioning financial systems reduce the burden on external financing, which hinder firm and industrial expansion (Mordi 2010, p.8). Role played by the financial sector in the economy A crucial role that the financial sector plays within an economy in the facilitation of growth is the reduction of enforcement, information, and transaction costs. This function is achieved through a number of unique functions performed by the financial sector. Therefore, the financial sector has some basic functions that include provision of efficient modes of payment for the whole economy, and intermediary between lenders and borrowers. The

Symbolic Learning Methods Essay Example for Free

Symbolic Learning Methods Essay Abstract In this paper, performance of symbolic learning algorithms and neural learning algorithms on different kinds of datasets has been evaluated. Experimental results on the datasets indicate that in the absence of noise, the performances of symbolic and neural learning methods were comparable in most of the cases. For datasets containing only symbolic attributes, in the presence of noise, the performance of neural learning methods was superior to symbolic learning methods. But for datasets containing mixed attributes (few numeric and few nominal), the recent versions of the symbolic learning algorithms performed better when noise was introduced into the datasets. 1. Introduction The problem most often addressed by both neural network and symbolic learning systems is the inductive acquisition of concepts from examples [1]. This problem can be briefly defined as follows: given descriptions of a set of examples each labeled as belonging to a particular class, determine a procedure for correctly assigning new examples to these classes. In the neural network literature, this problem is frequently referred to as supervised or associative learning. For supervised learning, both the symbolic and neural learning methods require the same input data, which is a set of classified examples represented as feature vectors. The performance of both types of learning systems is evaluated by testing how well these systems can accurately classify new examples. Symbolic learning algorithms have been tested on problems ranging from soybean disease diagnosis [2] to classifying chess end games [3]. Neural learning algorithms have been tested on problems ranging from converting text to speech [4] to evaluating moves in backgammon [5]. In this paper, the current problem is to do a comparative evaluation of the performances of the symbolic learning methods which use decision trees such as ID3 [6] and its revised versions like C4.5 [7] against neural learning methods like Multilayer perceptrons [8] which implements a feed-forward neural network with error back propagation. Since the late 1980s, several studies have been done that compared the performance of symbolic learning approaches to the neural network techniques. Fisher and McKusick [9] compared ID3 and Backpropagation on the basis of both prediction accuracy and the length of training. According to their conclusions, Backpropagation attained a slightly higher accuracy. Mooney et al., [10] found that ID3 was faster than a Backpropagation network, but the Backpropagation network was more adaptive to noisy data sets. Shavlik et al., [1] compared ID3 algorithm with perceptron and backpropagation neural learning algorithms. They found that in all cases, backpropagation took much longer to train but the accuracies varied slightly depending on the type of dataset. Besides accuracy and learning time, this paper investigated three additional aspects of empirical learning, namely, the dependence on the amount of training data, the ability to handle imperfect data of various types and the ability to utilize distributed output encodings. Depending upon the type of datasets they worked on, some authors claimed that symbolic learning methods were quite superior to neural nets while some others claimed that accuracies predicted by neural nets were far better than symbolic learning methods. The hypothesis being made is that in case of noise free data, ID3 gives faster results whose accuracy will be comparable to that of back propagation techniques. But in case of noisy data, neural networks will perform better than ID3 though the time taken will be more in case of neural networks. Also, in the case of noisy data, performance of C4.5 and neural nets will be comparable since C4.5 too is resistant to noise to an extent due to pruning. 2. Symbolic Learning Methods In ID3, the system constructs a decision tree from a set of training objects. At each node of the tree the training objects are partitioned by their value along a single attribute. An information theoretic measure is used to select the attribute whose values improve prediction of class membership above the accuracy expected from a random guess. The training set is recursively decomposed in this manner until no remaining attribute improves prediction in a statistically significant manner when the confidence factor is supplied by the user. So, ID3 method uses Information Gain heuristic which is based on Shannon’s entropy to build efficient decision trees. But one dis advantage with ID3 is that it overfits the training data. So, it gives rise to decision trees which are too specific and hence this approach is not noise resistant when tested on novel examples. Another disadvantage is that it cannot deal with missing attributes and requires all attributes to have nominal values. C4.5 is an improved version of ID3 which prevents over-fitting of training data by pruning the decision tree when required, thus making it more noise resistant. 3. Neural Network Learning Methods Multilayer perceptron is a layered network comprising of input nodes, hidden nodes and output nodes [11]. The error values are back propagated from the output nodes to the input nodes via the hidden nodes. Considerable time is required to build a neural network but once it is done, classification is quite fast. Neural networks are robust to noisy data as long as too many epochs are not considered since they do not overfit the training data. 4. Evaluation Design For the evaluation purposes, a free and popular software tool called Weka (Waikato Environment for Knowledge Acquisition) is used. This software has the implementations of several machine learning algorithms made easily accessible to the user with the help of graphical user interfaces. The training and the test datasets have been taken from the UCI machine learning repository. Two different types of datasets will be used for the evaluation purposes. One type of datasets contain only symbolic attributes (Symbolic Datasets) and the other type contain mixed attributes (Numeric Datasets). Performance of the different learning methods will be evaluated using the original datasets which do not contain any noise and after introducing noise into them. Noise is introduced in the class attributes of the datasets by using the ‘AddNoise’ filter option in Weka which adds the specified percentage of noise randomly into the datasets. Symbolic Datasets are those which contain only symbolic attributes. Symbolic learning methods like ID3 and its recent developments can be run only on datasets where all the attributes are nominal. In Weka, these nominal attributes are automatically converted to numeric ones for neural network learning methods. So, preprocessing is not required in this type of datasets. Numeric Datasets are those which contain few nominal and few numeric attributes. Since symbolic learning methods like ID3 and its recent developments can be run only on datasets where all the attributes are nominal, these datasets first need to be preprocessed. A ‘Discretize’ filter option available in Weka is used to discretize all the non-symbolic attribute values into individual intervals so that each attribute can now be treated as a symbolic one. Initially, the entire data being considered is randomized. Two types of evaluation techniques are being used to analyze the data. (a) Percentage Split: In general, the data will be split up randomly into training data and test data. In the experiments conducted, the data will be split such that training data comprises 66% of the entire data and the rest is used for testing. (b) K-fold Cross-validation: In general, the data is split into k disjoint subsets and one of it is used as testing data and the rest of them are used as training data. This is continued till every subset has been used once as a testing dataset. In the experiments conducted, 5-fold cross validation was done. 5. Experimental Results Experiments were conducted on two symbolic datasets and two numeric datasets. The two symbolic datasets are tic-tac-toe and chess. The two numeric datasets are segment and teacher’s assistant evaluation (tae). DataSet 1 : TIC-TAC-TOE (a) 5-fold cross validation (i)Without any noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii) Percentage of noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.03 6.16 0.02 0.06 0.01 % correct 67.4322 81.8372 75.8873 73.5908 71.2944 % incorrect 28.0793 18.1628 24.1127 26.4092 28.7056 % not classified 4.4885 0 0 0 0 Time to build 0.06 6.35 0.06 0.01 0.02 % correct 86.1169 97.4948 85.8038 87.5783 83.1942 % incorrect 11.691 2.5052 14.1962 12.4217 16.8058 % not classified 2.1921 0 0 0 0 (b) Percentage split with training data being 66% and the rest is testing data (i)Without Noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii)Percentage of Noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.05 6.5 0.01 0.01 0.02 % correct 85.5828 97.546 83.1288 88.0368 82.2086 % incorrect 11.0429 2.454 16.8712 11.9632 17.7914 % not classified 3.3742 0 0 0 0 Time to build 0.04 6.15 0.02 0.02 0.01 % correct 68.4049 80.6748 73.9264 72.3926 71.4724 % incorrect 28.2209 19.3252 26.0736 27.6074 28.5276 % not classified 3.3742 0 0 0 0 For the tic-tac-toe dataset, in the presence of noise, neural nets had better prediction accuracies than all the other algorithms as expected. Though C4.5 gives better accuracy than ID3, its accuracy is still lower in comparison to Neural Nets. If the pruning factor (confidence factor was lowered) was increased, the prediction accuracies of C4.5 dropped a little. But in the absence of noise, the performances of ID3 and Multilayer Perceptron should have been comparable. But the performance of Multilayer Perceptron is quite superior to ID3. DataSet 2 : CHESS (a) 5-fold cross validation (i)Without any noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii) Percentage of noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.36 47.75 0.21 0.18 0.19 % correct 81.1952 86.796 89.0488 84.6683 88.4856 % incorrect 18.8048 13.204 10.9512 15.3317 11.5144 % not classified 0 0 0 0 0 Time to build 0.21 47.67 0.15 0.05 0.1 % correct 99.562 97.4656 99.3742 99.3116 99.2178 % incorrect 0.438 2.5344 0.6258 0.6884 0.7822 % not classified 0 0 0 0 0 (b) Percentage split with training data being 66% and the rest is testing data (i)Without Noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii)Percentage of Noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.33 41.73 0.24 0.19 0.19 % correct 80.1288 85.7406 87.5805 82.6127 87.6725 % incorrect 19.8712 14.2594 12.4195 17.3873 12.3275 % not classified 0 0 0 0 0 Time to build 0.13 43.55 0.06 0.06 0.08 % correct 99.448 97.1481 99.08 98.988 99.08 % incorrect 0.552 2.8519 0.92 1.012 0.92 % not classified 0 0 0 0 0 For the chess dataset, in the absence of noise, the performance of ID3 is better than that of Multilayer perceptron and takes lesser time. For the noisy data, back propagation predicts better accuracies than that of ID3 as expected, but the performance of C4.5 is slightly higher than back propagation. The reason for this could be that the feature space in this dataset is more relevant. So, C4.5 builds a tree and prunes it to get a more efficient tree. DataSet 3 : SEGMENT (a) 5-fold cross validation (i) Without any noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii) Percentage of noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.07 9.64 0.04 0.04 0.03 % correct 68.9333 80.8667 81.2667 79.6 80.5333 % incorrect 21.3333 19.1333 18.7333 20.4 19.4667 % not classified 9.7333 0 0 0 0 Time to build 0.05 10.3 0.02 0.23 0.12 % correct 88.0667 90.6 91.6 94 94.3333 % incorrect 5.2 9.4 8.4 6 5.6667 % not classified 6.7333 0 0 0 0 (b) Percentage split with training data being 66% and the rest is testing data (i) Without Noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii) Percentage of Noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.07 11.73 0.03 0.04 0.03 % correct 72.9412 82.549 82.1569 82.549 81.3725 % incorrect 19.6078 17.451 17.8431 17.451 18.6275 % not classified 7.451 0 0 0 0 Time to build 0.06 9.87 0.03 0.02 0.03 % correct 89.8039 87.6471 92.1569 93.7255 90.1961 % incorrect 4.1176 12.3529 7.8431 6.2745 9.8039 % not classified 6.0784 0 0 0 0 Segment, being a numeric dataset, all the attribute values had to be discretized before running the algorithms. In the absence of noise, ID3 performs slightly better than back propagation and the performance of J48 (implementation of C4.5 in Weka) is much better than ID3 and backpropagation. But a very interesting observation was found. In the absence of noise, the performance of an unpruned tree generated by C4.5 was quite superior to the rest. In the presence of noise, the performances of back propagation and C4.5 were comparable. DataSet 4 : TAE (a) 5-fold cross validation (i) Without any noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii) Percentage of noisy data = 10% Time to % % build correct incorrect ID3 0.02 53.6424 37.0861 Multilayer Perceptron 0.16 38.4106 61.5894 J48 0.02 52.9801 47.0199 C4.5 unpruned 0.01 56.2914 43.7086 C4.5 confidence factor = 0.1 0.01 54.3046 45.6954 (b) Percentage split with training data being 66% and the rest is testing data (i) Without Noise: Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 (ii) Percentage of Noisy data = 10% Classifiers ID3 Multilayer Perceptron J48 C4.5 unpruned C4.5 confidence factor = 0.1 Time to build 0.01 0.17 0.01 0.01 0.01 % correct 38.4615 44.2308 44.2308 50 44.2308 % incorrect 40.3846 55.7692 55.7692 50 55.7692 % not classified 21.1538 0 0 0 0 Time to build 0.02 2.23 0.03 0.02 0.01 % correct 44.2308 57.6923 51.9231 55.7692 42.3077 % incorrect 34.6154 42.3077 48.0769 44.2308 57.6923 % not classified 21.1538 0 0 0 0 Classifiers % not classified 0 0 0 0 0 Time to build 0.02 0.18 0.02 0.01 0.01 % correct 54.3046 54.9669 48.3444 50.9934 47.0199 % incorrect 35.0993 45.0331 51.6556 49.0066 52.9801 % not classified 10.596 0 0 0 0 TAE, being a numeric dataset, its attribute values had to be discretized too before running the algorithms. But after observing the results, it is very clear that the random discretization provided by Weka did not generate good intervals due to which the overall accuracy predicted by all the methods is quite poor. Again, interestingly an unpruned tree built by C4.5 seems to give high prediction accuracies relative to the rest in most of the cases. In this case, for cross-validation approach and noisy data, surprisingly the performance of back-propagation was very poor. One reason for this could be that only few epochs of the training data were run to build the neural network. In the absence of noise, accuracy prediction of Multilayer perceptron was either comparable or greater than that of ID3. 6. Conclusion No single machine learning algorithm can be considered superior to the rest. The performance of each algorithm depends on what type of dataset is being considered, whether the f eature space is relevant and whether the data contains noise. In the absence of noise, in some cases, the performance of ID3 was comparable or sometimes better than back-propagation and was faster but in some cases Multilayer perceptron performed better. When noisy datasets were considered, back propagation definitely did better than ID3 though it took more time to build the neural network. But in the presence of noise, in some cases, C4.5 gave faster and better results when the attributes being considered were relevant. But some surprising observations were made when the attribute values of the numeric datasets were discretized, the prediction accuracy of an unpruned tree generated by C4.5 algorithm was much higher than the rest. This shows that the unpruned tree generated by C4.5 is not the same as that generated by ID3. References: 1.Mooney, R., Shalvik, J., and Towell, G. (1991): Symbolic and Neural Learning Algorithms An experimental comparison, in Machine Learning 6, pp. 111-143. 2. Michalski, R.S., Chilausky, R.L. (1980): Learning by being told and learning from examples An experimental comparison of two methods of knowledge acquisition in the context of developing an expert system for soybean disease diagnosis, in Policy Analysis and Information Systems, 4, pp. 125-160. 3. Quinlan, J.R. (1983): Learning efficient classification procedures and their application to chess end games in R.S. Michalski, J.G. Carbonell, T.M. Mitchell (Eds.), in Machine learning: An artificial intelligence approach (Vol. 1). Palo Alto, CA: Tioga. 4. Sejnowski, T.J., Rosenberg, C. (1987): Parallel networks that learn to pronounce English text, in Complex Systems, 1, pp. 145-168. 5. Tesauro, G., Sejnowski, T.J. (1989): A p arallel network that learns to play backgammon, in Artificial Intelligence, 39, pp. 357-390. 6. Quinlan, J.R. (1986): Induction on Decision Trees, in Machine Learning 1, 1 7. Quinlan, J.R. (1993): C4.5 – Programs for Machine Learning. San Mateo: Morgan Kaufmann. 8. Rumelhart, D., Hinton, G., Williams, J. (1986): Learning Internal Representations by Error Propagation, in Parallel Distributed Processing, Vol. 1 (D. Rumelhart k J. McClelland, eds.). MIT Press. 9. Fisher, D.H. and McKusick, K.B. (1989): An empirical comparison of ID3 and backpropagation, in Proc. of the Eleventh International Joint Conference on Artificia1 Intelligence (IJCAI-89), Detroit, MI, August 20-25, pp. 788-793. 10. Mooney, R., Shavlik, J., Towell, G., and Gove, A.(1989): An experimental comparison of symbolic and connectionist learning algorithms, in Proc. of the Eleventh International Joint Conference on Artificial Intelligence (IJCAI-89), Detroit, MI, August 20-25, pp. 775-780. 11. McClelland, J. k Rumelhart, D. (1988). Explorations in Parallel Distributed Processing, MIT Press, Cambridge, MA.